EP1633234A2 - Systemes et procedes permettant de determiner la pression intracranienne de fa on non invasive et ensembles de transducteurs acoustiques destines a etre utilises dans ces systemes - Google Patents

Systemes et procedes permettant de determiner la pression intracranienne de fa on non invasive et ensembles de transducteurs acoustiques destines a etre utilises dans ces systemes

Info

Publication number
EP1633234A2
EP1633234A2 EP04754563A EP04754563A EP1633234A2 EP 1633234 A2 EP1633234 A2 EP 1633234A2 EP 04754563 A EP04754563 A EP 04754563A EP 04754563 A EP04754563 A EP 04754563A EP 1633234 A2 EP1633234 A2 EP 1633234A2
Authority
EP
European Patent Office
Prior art keywords
acoustic
data
icp
abp
transducer
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
EP04754563A
Other languages
German (de)
English (en)
Other versions
EP1633234A4 (fr
Inventor
Pierre Mourad
Brandt Mohr
Michel Kliot
Robert C. A. Frederickson
Lee R. Thompson
Jason Seawall
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Frederickson Robert C A
Kliot Michel
MOHR, BRANDT
Mourad Pierre
Seawall Jason
THOMPSON, LEE R.
University of Washington
PhysioSonics Inc
Original Assignee
Allez Physionix Ltd
University of Washington
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Allez Physionix Ltd, University of Washington filed Critical Allez Physionix Ltd
Priority to EP11177783A priority Critical patent/EP2392262A1/fr
Publication of EP1633234A2 publication Critical patent/EP1633234A2/fr
Publication of EP1633234A4 publication Critical patent/EP1633234A4/fr
Ceased legal-status Critical Current

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/06—Measuring blood flow
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/03—Detecting, measuring or recording fluid pressure within the body other than blood pressure, e.g. cerebral pressure; Measuring pressure in body tissues or organs
    • A61B5/031—Intracranial pressure
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/40—Detecting, measuring or recording for evaluating the nervous system
    • A61B5/4058—Detecting, measuring or recording for evaluating the nervous system for evaluating the central nervous system
    • A61B5/4064—Evaluating the brain
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/44—Constructional features of the ultrasonic, sonic or infrasonic diagnostic device
    • A61B8/4444—Constructional features of the ultrasonic, sonic or infrasonic diagnostic device related to the probe
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/44—Constructional features of the ultrasonic, sonic or infrasonic diagnostic device
    • A61B8/4483—Constructional features of the ultrasonic, sonic or infrasonic diagnostic device characterised by features of the ultrasound transducer
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
    • A61B8/0808—Detecting organic movements or changes, e.g. tumours, cysts, swellings for diagnosis of the brain
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/44—Constructional features of the ultrasonic, sonic or infrasonic diagnostic device
    • A61B8/4444—Constructional features of the ultrasonic, sonic or infrasonic diagnostic device related to the probe
    • A61B8/4472—Wireless probes
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/48—Diagnostic techniques
    • A61B8/485—Diagnostic techniques involving measuring strain or elastic properties
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/58—Testing, adjusting or calibrating the diagnostic device
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients

Definitions

  • the present invention relates to methods and systems for determining intracranial pressure (ICP) based on variable physiological parameters that can be measured using non-invasive or minimally invasive techniques.
  • the present invention relates to methods and systems for acquiring and processing acoustic data to derive accurate ICP determinations non-invasively.
  • the present invention provides methods and systems for locating target areas based on their acoustic properties and for acoustic scanning of an area, identification of a target site within the area of interest based on its acoustic properties, and automated focusing of an acoustic source and/or detector at the target site.
  • Acoustic transducer assemblies, arrays and source/detector combinations for use in methods and systems of the present invention are also provided.
  • Ultrasound imaging is a non-invasive, diagnostic modality that provides information relating to tissue properties and spatial location of physiological structures.
  • ultrasound may be used in various modes to produce images of objects or structures within a patient, hi a transmission mode, an ultrasound transmitter is placed on one side of an object and the sound is transmitted through the object to an ultrasound receiver.
  • An image may be produced in which the brightness of each image pixel is a function of the amplitude of the ultrasound that reaches the receiver (attenuation mode), or the brightness of each pixel may be a function of the time required for the sound to reach the receiver (time-of- flight mode).
  • an image may be produced in which the pixel brightness is a function of the amplitude of reflected ultrasound (reflection or backscatter or echo mode), h a Doppler mode of operation, the tissue (or object) is imaged by measuring the phase shift of the ultrasound reflected from the tissue (or object) back to the receiver.
  • Ultrasonic transducers for medical applications are constructed from one or more piezoelectric elements activated by electrodes. Such piezoelectric elements may be constructed, for example, from lead zirconate titanate (PZT), polyvinylidene diflouride (PVDF), PZT ceramic/polymer composite, and the like.
  • the electrodes are connected to a voltage source, a voltage waveform is applied, and the piezoelectric elements change in size at a frequency corresponding to that of the applied voltage.
  • a voltage waveform is applied, the piezoelectric elements emit an ultrasonic wave into the media to which it is coupled at the frequencies contained in the excitation waveform.
  • an ultrasonic wave strikes the piezoelectric element, the element produces a corresponding voltage across its electrodes.
  • Numerous ultrasonic transducer constructions are known in the art.
  • ultrasonic transducers When used for imaging, ultrasonic transducers are provided with several piezoelectric elements arranged in an array and driven by different voltages. By controlling the phase and amplitude of the applied voltages, ultrasonic waves combine to produce a net ultrasonic wave that travels along a desired beam direction and is focused at a selected point along the beam. By controlling the phase and the amplitude of the applied voltages, the focal point of the beam can be moved in a plane to scan the subject. Many such ultrasonic imaging systems are well known in the art.
  • Doppler ultrasound has been in use in medicine for many years. Doppler ultrasound techniques measure the frequency shift (the "Doppler Effect") of reflected sound, which indicates the velocity of the reflecting material. Long-standing applications of Doppler ultrasound include monitoring of the fetal heart rate during labor and delivery and evaluating blood flow in the carotid artery. The use of Doppler ultrasound has expanded greatly in the past two decades, and Doppler ultrasound is now used in many medical specialties, including cardiology, neurology, radiology, obstetrics, pediatrics, and surgery. Doppler technology today allows detection of flow in intracranial arteries. Transcranial Doppler (TCD) techniques require application of the ultrasound to those areas of the skull where the bone is relatively thin.
  • TCD Transcranial Doppler
  • the frequency of the Doppler signal is also adjusted, and pulsed wave rather than continuous wave ultrasound is used to augment the transmission of ultrasound waves through the skull.
  • Velocities from the cerebral arteries, the internal carotids, the basilar and the vertebral arteries can be sampled by altering the transducer location and angle, and the instrument's depth setting.
  • the most common windows in the cranium are located in the orbit (of the eye), and in the temporal and suboccipital regions.
  • TCD ultrasonography provides an easy-to-use, non-invasive, non-radioactive, and relatively inexpensive method to assess intracerebral hemodynamics with time resolution and provides reliable detection of cerebral perfusion changes.
  • cerebrovascular responsiveness to various physiological and pharmacological challenges can be assessed instantaneously, and various cerebral circulatory tests can be repeated often and safely. Rapid changes of cerebral perfusion over time can be easily followed, documented and analyzed.
  • Intracranial Pressure Intracranial Pressure
  • Intracranial pressure Normal, healthy mammals, particularly humans, have a generally constant intracranial volume and, hence, a generally constant intracranial pressure.
  • Various conditions produce changes in the intracranial volume and, consequently, produce changes in intracranial pressure.
  • Increases in intracranial pressure may produce conditions under which the intracranial pressure rises above normal and approaches or even equals the mean arterial pressure, resulting in reduced blood flow to the brain.
  • Elevated intracranial pressure not only reduces blood flow to the brain, but it also affects the normal metabolism of cells within the brain. Under some conditions, elevated intracranial pressures may cause the brain to be mechanically compressed, and to herniate. The most common cause of elevated intracranial pressure is head trauma.
  • Additional causes of elevated intracranial pressure include shaken-baby syndrome, epidural hematoma, subdural hematoma, brain hemorrhage, meningitis, encephalitis, lead poisoning, Reye's syndrome, hypervitaminosis A, diabetic ketoacidosis, water intoxication, brain tumors, other masses or blood clots in the cranial cavity, brain abcesses, stroke, ADEM (acute disseminated encephalomyelitis), metabolic disorders, hydrocephalus, and dural sinus and venous thrombosis. Changes in intracranial pressure, particularly elevated intracranial pressure, are very serious and may be life threatening. They require immediate treatment and continued monitoring.
  • Conventional intracranial pressure monitoring devices include: epidural catheters; subaraclinoid bolt/screws; ventriculostomy catheters; and fiberoptic catheters. All of these methods and systems are invasive.
  • An epidural catheter may be inserted, for example, during cranial surgery.
  • the epidural catheter has a relative low risk of infection and it does not require transducer a djustment with head movement, but the accuracy of sensing decreases through dura, and it is unable to drain CSF.
  • the subarachnoid bolt/screw technique requires minimal penetration of the brain, it has a relatively low risk of infection, and it provides a direct pressure measurement, but it does require penetration of an intact skull and it poorly drains CSF.
  • the ventriculostomy catheter technique provides CSF drainage and sampling and it provides a direct measurement of intracranial pressure, but the risks of infection, intracerebral bleeding and edema along the cannula track are significant, and it requires transducer repositioning with head movement.
  • the fiber optic catheter technique is versatile because the catheter may be placed in the ventricle or in the subaraclinoid space or brain tissue, and it does not require adjustment of the transducer with head movement, but it requires a separate monitoring system.
  • U.S. P atent 5 ,951,477 o f Ragauskas et al. discloses an apparatus for non-invasively measuring intracranial pressure using an ultrasonic Doppler device that detects the velocities of the blood flow inside the optic artery for both intracranial and extracranial optic artery portions.
  • the eye in which the blood flow is monitored is subjected to a small pressure, which is sufficient to equalize the blood flow measurements of the intracranial and extracranial portions of the optic artery.
  • the pressure at which such equalization occurs is disclosed to be an acceptable indication of the intracranial pressure.
  • a pressurized chamber is sealed to the perimeter around an eye and the pressure in the chamber is controlled to equalize blood velocities of intracranial and extracranial portions of the optic artery.
  • U.S. Patent 5,388,583, to Ragauskas et al. discloses an ultrasonic non-invasive technique for deriving the time dependencies of characteristics of certain regions in the intracranial medium. Precise measurements of the transit travel times of acoustic pulses are made and processed to extract variable portions indicative of, for example, the pulsatility due to cardiac pulses of a basal artery or a cerebroventricle or the variation in the pressure of brain tissue, as well as changes in the cross-sectional dimension of the basal artery and ventricle. Frequency and phase detection techniques are also described.
  • U.S. Patent 5,411,028 to Bonnefous discloses an ultrasonic echograph used for the measurement of various blood flow and blood vessel parameters that provide information for calculating determinations relating to the elasticity or compliance of an artery and its internal pressure.
  • U.S. Patent 5,117,835 to Mick discloses a method and apparatus for non-invasively measuring changes in intracranial pressure by measuring changes in the natural frequency and frequency response spectrum of the skull bone. Changes in the natural frequency and frequency response spectrum of the skull are measured by applying a mechanical forced oscillation stimulus that creates a mechanical wave transmission through the bone, and then sensing the frequency response spectrum. Comparison of spectral response data over time shows trends and changes in ICP.
  • U.S. Patent 6,129,682 to Borchert et al. discloses a method for non-invasively determining ICP based on intraocular pressure (IOP) and a parameter of the optic nerve, such as thickness of the retinal nerve fiber layer or anterior-posterior position of the optic nerve head.
  • IOP intraocular pressure
  • U.S. Patent 6,086,533 to Madsen et al. discloses systems for non-invasive measurement of blood velocity based on the Doppler shift, and correlation of blood velocity before and after the manual application of an externally applied pressure, to provide a measure of intracranial pressure, ophthalmic pressure, and various other body conditions affecting blood perfusion.
  • U.S. Patent 5,919,144 to Bridger et al. discloses a non-invasive apparatus and method for measuring intracranial pressure based on the properties of acoustic signals that interacted with the brain, such as acoustic transmission impedance, resonant frequency, resonance characteristics, velocity of sound, and the like. Low intensity acoustic signals having frequencies of less than 100 kHz are used.
  • U.S. Patent 4,984,567 to Kageyama et al. discloses an apparatus for measuring intracranial pressure using ultrasonic waves. Data from interference reflection waves caused by multiple reflections of incident ultrasonic waves at the interstitial boundaries within the cranium are analyzed for frequency, and the time difference between the element waves of the interference reflection wave is calculated and provided as output.
  • the device described incorporates an electrocardiograph for detecting the heart beat, a pulser for generating a voltage pulse, an ultrasonic probe for receiving the pulse and transmitting an ultrasonic pulse into the cranium and receiving the echo of the incident wave, and a processor for making various calculations.
  • U.S. Patent 5,951,476 to Beach provides a method for detecting brain microhemorrhage by projecting bursts of ultrasound into one or both of the temples of the cranium, or into the medulla oblongata, with the readout of echoes received from different depths of tissue displayed on a screen.
  • the readouts of the echoes indicated accrued microshifts of the brain tissue relative to the cranium.
  • the timing of the ultrasound bursts is required to be synchronized with the heart pulse of the patient.
  • U.S. Patent 6,042,556 discloses a method for determining phase advancement of transducer elements in high intensity focused ultrasound. Specific harmonic echoes are distributed in all directions from the treatment volume, and the temporal delay in the specific harmonic echoes provides a measure of the propagation path transit time to transmit a pulse that converges on the treatment volume.
  • U.S. Patent 3,872,858 discloses an echo encephalo graph for use in the initial diagnosis of midline structure lateral shift that applies an ultrasonic pulse to a patient's head, the pulse traveling to a predetermined structure and being partially reflected as an echo pulse. Shifts are determined by measuring the travel time of the echo pulse.
  • U.S. Patent 4,984,567 describes an apparatus for measuring intracranial pressure based on the ultrasonic assay of changes in the thickness of the dura covering the brain induced by changes in ICP.
  • the reflected signals are processed, in a known manner, to generate an echo encephalogram (Echo EG), which is plotted as a function of amplitude vs. distance.
  • Echo EG echo encephalogram
  • a portion of the Echo EG signal is selected and integrated over the selected portion to generate an echo pulsograph (EPG) signal.
  • EPG echo pulsograph
  • the EPG signal is used to provide information regarding the physiological state of tissue, h one specific embodiment, the EPG signal is used to provide a quantitative measure of ICP using the relationship described at Col. 8, line 7.
  • PCT International Publication No. WO 02/43564 which is incorporated herein by reference in its entirety, discloses methods and systems for assessment of tissue properties, non-invasively, by acquiring data relating to at least one aspect of intrinsic and/or induced tissue displacement, or associated biological responses. Data relating to tissue displacement and associated biological changes are acquired by detecting acoustic properties of tissue using ultrasound interrogation pulses, preferably in a scatter or Doppler detection mode. Specific applications for such systems and methods include noninvasive assessment and monitoring of ICP, arterial blood pressure (ABP), CNS autoregulation status, vasospasm, stroke, local edema, infection and vasculitus, as well as diagnosis and monitoring of diseases and conditions that are characterized by physical changes in tissue properties.
  • ICP arterial blood pressure
  • ABSP arterial blood pressure
  • CNS autoregulation status vasospasm
  • stroke stroke
  • local edema infection and vasculitus
  • NASA has also worked on the development of methods and systems for noninvasive intracranial pressure measurement. Intracranial pressure dynamics are important for understanding adjustments to altered gravity. ICP may be elevated during exposure to microgravity conditions. S ymptoms o f space adaptation syndrome are similar to those of elevated intracranial pressure, including headache, nausea and projectile vomiting. The hypothesis that ICP is altered in microgravity environments is difficult to test, however, as a result of the invasive nature of conventional ICP measurement techniques. NASA has therefore developed a modified pulsed phase-locked loop (PPLL) method for measuring ICP based on detection of skull movements which occur with fluctuations in ICP. Detection of skull pulsation uses an ultrasound technique in which slight changes in the distance between an ultrasound transducer and a reflecting target are measured.
  • PPLL pulsed phase-locked loop
  • the instrument transmits a 500kHz ultrasonic tone burst through the cranium, which passes through the cranial cavity, reflects off the inner surface of the opposite side of the skull, and is received by the same transducer.
  • the instrument compares the phase of emitted and received waves and alters the frequency of the next stimulus to maintain a 90 degree phase difference between the ultrasound output and the received signal.
  • Experimental data demonstrated that the PPLL output was highly and predictably related to directly measured ICP.
  • U.S. Patent Applications 2001/0039386 Al and 2002/0183650 Al disclose methods for e liminating s low d rift a rtifacts from s onomicrometer s ignals t o i mprove t he q uality o f ICP measurement data obtained from skull diameter measurements. These methods involve using a neural network or another non-linear engine to extract a heartbeat component from the sonomicrometer output.
  • ABP Arterial blood pressure
  • ABP is most commonly measured non-invasively using a pneumatic cuff, often described as pneumatic plethysmography or Kortkoff s method. While this mode of measurement i s s imple and i nexpensive to p erfortn, i t d oes n ot p rovide t he m ost a ccurate measure of ABP, and it is susceptible to artifacts resulting from the condition of arterial wall, the s ize o f t he p atient, t he h emodynamic s tatus of t he p atient, a nd a utonomic t one o f t he vascular smooth muscle.
  • Methods and systems of the present invention provide accurate assessment and monitoring of ICP based on one or more variable physiological parameters that can be measured using non-invasive or minimally invasive techniques.
  • One of the variable physiological parameters may relate to intracranial blood flow and may, for example, be quantified as an acoustic property of tissue or blood that is related to intracranial blood flow or intracranial flow velocity, such as cerebral blood flow or flow velocity.
  • ICP is determined based on an acoustic property of cerebral tissue, or on blood flow or cerebral blood flow velocity, and/or arterial blood pressure (ABP).
  • patient ICP is determined based on at least two variable parameters: (1) acoustic scatter or flow velocity in the middle cerebral artery (V mca) measured, for example, using a TCD device; and (2) ABP measured invasively or non-invasively.
  • ABP may be measured using conventional techniques, or using ultrasound techniques as described herein, h one embodiment, ABP is measured non- invasively, using "active" and/or "passive" ultrasound techniques, in a cranial blood vessel such as the MCA or a carotid or vertebral artery.
  • V_mca may be determined simultaneously or alternatively with ABP using an ultrasound device of the present invention.
  • patient ICP is determined based on: (1) D oppler or other acoustic measurements, such as acoustic scatter, taken from a target site on or within or in proximity to a cranial blood vessel such as the MCA, a carotid artery, or another cranial blood vessel; and (2) ABP and/or CNS tissue displacement measured using active and/or passive acoustic techniques described herein in one or more target CNS sites different from the target site on or within or in proximity to a cranial blood vessel.
  • D oppler or other acoustic measurements such as acoustic scatter
  • a target site on or within or in proximity to a cranial blood vessel such as the MCA, a carotid artery, or another cranial blood vessel
  • ABP and/or CNS tissue displacement measured using active and/or passive acoustic techniques described herein in one or more target CNS sites different from the target site on or within or in proximity to a cranial blood vessel.
  • patient JCP is determined based on: (1) acoustic scatter data acquired from a target site on or within or in proximity to a cranial blood vessel, such as the MCA or a carotid or a vertebral artery; and (2) ABP and/or acoustic scatter data acquired from one or more target CNS sites different from the target site on or within or in proximity to a cranial blood vessel.
  • a cranial blood vessel such as the MCA or a carotid or a vertebral artery
  • CNS tissue stiffness tissue stiffness
  • endogenous and/or induced tissue displacement partial pressure of gases associated with brain respiration and metabolism
  • pCO 2 partial pressure of gases associated with brain respiration and metabolism
  • blood perfusion hematocrit
  • EKG electrophysiological properties of tissue, such as evoked potentials
  • electrophysiological properties of tissue such as evoked potentials
  • one or more of these physiological properties of blood and/or CNS tissue is used, together with acoustic scatter data and/or Doppler analysis of scatter data, and/or ABP, to determine instantaneous ICP.
  • one or more physiological properties of blood and/or CNS tissue may be used, without requiring data relating to flow velocity, to determine instantaneous ICP.
  • Acoustic properties of tissues including blood, blood vessel walls and blood vessels, and tissue displacement, may be evaluated using ultrasound techniques described in PCT International Publication WO 02/43564 and U.S. Patent Application No. US 2002/0095087 Al, which are incorporated herein by reference in their entireties.
  • acoustic (ultrasound) techniques are used to acquire data relating to intrinsic (endogenous) tissue displacement
  • acoustic (ultrasound) techniques are used to stimulate or probe target tissue, or induce a response at a target tissue site, by the application of focused ultrasound.
  • acoustic (ultrasound) techniques use the application of focused ultrasound to produce oscillation of targeted tissue.
  • Ultrasound backscatter and/or emission data are related to intrinsic tissue displacements, which can be related to ICP, ABP, CPP, autoregulation status and various tissue properties and physiological conditions, hi some embodiments of methods and systems of the present invention, both passive and active ultrasound techniques may be used, simultaneously or alternatively, to assess tissue properties.
  • passive ultrasound techniques such as TCD techniques
  • passive and/or “active” ultrasound techniques are used to measure ABP.
  • the V_mca and ABP measurements may be used, for example, to determine ICP.
  • the magnitude and/or amplitude and/or phase of acoustic scatter from target tissue sites in the CNS undergoing intrinsic displacements during the course of arterial blood flow and CSF supply is directly related to the stiffness, e.g. Young's modulus, of the CNS tissue, and is therefore empirically related to ICP.
  • stiffness e.g. Young's modulus
  • relationships between the major and minor intrinsic oscillations of CNS tissue within a cardiac cycle, or within a cardiac cycle as modulated by one or more respiratory cycles are empirically related to ICP.
  • Additional properties of the intrinsic tissue displacement that may be determined and related to tissue properties include: various components of amplitude, such as maximum amplitude within a cardiac cycle, the ratio of the maximum amplitude to that of the mean or variance of subsequent oscillations within a cardiac cycle, all possible rates of change of intrinsic CNS tissue displacement or relaxation, such as the velocity or acceleration of displacement, and the like.
  • Additional data such as ABP measurements and/or respiration data, may be collected and used, with the acoustic data, to make various assessments and determinations of ICP, CPP, autoregulation status or capacity, and the like.
  • methods and systems of the present invention stimulate or probe target tissue, or induce a response at a target tissue site, by ' application of focused ultrasound.
  • the response of the targeted tissue to the application of focused ultrasound may be displacement or a change in relative position, a sensation such as pain, a change in temperature, a change in blood flow, or another detectable response.
  • application of an acoustic radiation force to "palpate" a target tissue location may be accomplished by administering one or more acoustic signals.
  • Non-invasive techniques such as ultrasound, optical techniques such as near infrared spectroscopy and optical coherence tomography, and other techniques, including magnetic resonance techniques, external electrophysiological stimulation, patient response, and the like are used to assess at least one response to the application of focused ultrasound.
  • a visualization or imaging technique such as ultrasound imaging or magnetic resonance imaging, may also be employed to assist in targeting the focused ultrasound pulse(s) and to assist in differentially localizing responsive tissues.
  • Biological materials such as CNS tissue, absorb some of the ultrasound as it propagates into and through the material. See, e.g., Rudenko et al. (1996), "Acoustic radiation force and streaming induced by focused non-linear ultrasound in a dissipative medium," J. Acoust. Soc. Am 99(5) 2791-2798. Also, at the boundaries between different tissue types, such as between CSF and brain tissue, there is an 'impedance mismatch' (that is, differences between the product of density and speed of sound from one tissue to another) that allows ultrasound to push on the interface. See, e.g., Chu and Apfel (1982) "Acoustic radiation pressure produced by a beam of sound," J. Acoust. Soc.
  • Tissue displacement may thus be induced, and tissue may be acoustically palpated or oscillated, to produce displacement and other biological responses, and acoustic emissions, by application of focused ultrasound.
  • tissue may be acoustically palpated or oscillated, to produce displacement and other biological responses, and acoustic emissions, by application of focused ultrasound.
  • a single frequency acoustic source causes materials that are at least somewhat compliant, such as brain tissue, to move in a single direction relative to the source during propagation, while the material returns to its original location when propagation from the acoustic source is discontinued. Repeated pulses induce a repeated series of displacements and relaxations of the tissue.
  • one or more acoustic transducer(s) is placed in contact with or in proximity to a subject's skull.
  • An initial environmental assessment described below and preferably employing ultrasound techniques, may be made, if desired, to assess the characteristics of the environment between the acoustic source and the target tissue site, so that the magnitude of the acoustic force applied to the target tissue may be determined.
  • Environmental factors such as the distance between the acoustic transducer and various structural landmarks, such as the brain surface, the thickness of the skull, the thickness of the dura matter, the thickness of the arachnoid layer containing CSF, impedance mismatches between the various structures and tissues, and the like, may be determined.
  • the initial environmental assessment is determinative of various method and system parameters.
  • Environmental assessments may additionally be updated at intervals throughout a diagnostic or monitoring procedure.
  • an acoustic force is applied by an acoustic transducer, at a predete ⁇ nined frequency, to displace the brain tissue at a desired location, such as at the surface of the brain.
  • the deformation may be produced at any desired location within tissue, depending on the focus (foci) of the ultrasonic transducer(s) producing the acoustic radiation force.
  • variable foci ultrasonic transducers are provided, and a diagnostic procedure is carried out using a plurality of target tissue sites.
  • the focus (foci) of the ultrasonic transducer(s) is preferably provided in proximity to the cortical surface or a small distance below the cortical surface, to rn aximize t he t issue d isplacement i nduced b y t he radiation pressure t hat a rises from the impedance mismatch between brain and CSF or between brain and bone (depending on the frequency of the applied ultrasound). It is important to note, again, that the methods and systems of the present invention do not require the radiation force arising from the impedance mismatch described by Chu and Apfel to be significantly greater than that described by Rudenko et al.
  • the applied acoustic radiation force is sufficient to induce a detectable displacement in the CNS tissue, or the applied ultrasound beam is sufficient to produce a detectable biological response, without producing any medically undesirable changes in the examined tissue.
  • the acoustic radiation force applied must not produce shear in tissues in proximity to the target tissue of a magnitude sufficient to tear or damage tissue.
  • the applied ultrasound moreover, must not appreciably increase the temperature of examined tissue to the point of causing unacceptable damage, and it must not induce extensive or damaging cavitation or other sources of deleterious mechanical effects in the examined tissue.
  • Suitable ultrasound dosages may be determined using well known techniques. For example, Fry et al. studied the threshold ultrasonic dosages causing structural changes in mammalian brain tissue and illustrate, in their Fig.
  • the acoustic frequency must be low enough to penetrate the skull and high enough to produce measurable deformation in the target tissue at the location of interest. Within the parameters outlined above, higher frequency acoustic waves are more easily focused and, therefore, preferred.
  • the intensity must be high enough to deform the tissue, but not be so great as to induce undesirable changes in the examined tissue.
  • the pulse length is preferably relatively short, but long enough to create a measurable deformation or oscillation of the target tissue, as desired, while the pulse repetition frequency must be large enough to resolve medically interesting temporal features in the tissue, without inducing medically unacceptable changes in the tissue.
  • At least one acoustic property related to tissue displacement, or an associated biological response is determined and related to a tissue property and, ultimately, to a clinically important parameter.
  • the magnitude, or amplitude, of the displacement induced by the known acoustic force is directly related to the elasticity (or stiffness or compliance, e.g., Young's modulus) of the CNS tissue, and can therefore be empirically related to ICP.
  • Additional properties of the target tissue displacement that may be d etermined and related to tissue properties include: v arious components o f amplitude, such as maximum amplitude in the direction of the acoustic force or maximum amplitude perpendicular to the direction of acoustic force; all possible rates of change of the displacement or subsequent relaxation of the tissue, such as the velocity or acceleration of displacement or relaxation; the amplitude or rates of change of various components of the shape of the displacement; changes in Fourier or wavelett representations of the acoustic scatter signal associated with the displacement; properties of shear waves generated by the acoustic radiation force; properties of induced second harmonic deformation(s), and the like. Time displacements of pulse echoes returning from the target tissue are also indicative of the displacement amplitude and may be determined. These properties are all referred to as measures of "displacement.”
  • a second "active" mode of operation application of focused ultrasound produces oscillation of targeted tissue, and data relating to the acoustic signals emitted from the targeted tissue are collected. These signals are referred to herein as acoustic emissions, hi general, methods and systems of the present invention that relate to application of focused ultrasound may be used to produce oscillation of targeted tissue, and emitted acoustic signals are related to tissue properties and physiological conditions.
  • methods and systems of the present invention employ a confocal acoustic system comprising at least two acoustic transducers, driven at different frequencies, or a focal acoustic system comprising a single acoustic transducer driven at a given pulse repetition frequency (PRF), to induce an oscillatory radiation force in the target tissue, such as brain tissue.
  • the resulting oscillation is at a frequency that is the difference of the applied frequencies, at the target location that is marked by the overlap of the two confocal acoustic beams or, for the single transducer case, at the PRF.
  • D uring and after the application o f focused ultrasound the targeted tissue emits acoustic signals related to its intrinsic properties.
  • the second, active mode of operation may therefore be used to characterize tissue. Diagnostic ultrasound techniques may be used to measure the frequency or other properties of the emitted acoustic signal, which are empirically related to tissue properties.
  • Data relating to acoustic scatter produced as a result of focusing an ultrasound source on, or in, or " in proximity to, an intracranial blood vessel may be acquired using non-invasive means and may provide data for determining ICP using methodologies and systems of the present invention.
  • Properties of intracranial blood flow, such as flow velocity may also be determined using Transcranial Doppler (TCD) techniques, and may provide data for determining ICP.
  • TCD Transcranial Doppler
  • Methods and systems of the present invention may use raw acoustic scatter data from a target site within the CNS or within, on or in proximity to (collectively, "on") a cranial blood vessel such as the MCA, or tissue in proximity to such a target site, or may use processed acoustic data, such as Doppler data.
  • Acoustic properties of tissue may be determined, for example, by collecting acoustic scatter d ata u sing an ultrasound transducer aimed at, or having a focus on an intracranial blood vessel, and/or at another target site.
  • the target site is preferably a CNS tissue site, such as a cranial blood vessel, brain tissue, or the like.
  • the target site may be any tissue site (including blood, cerebral spinal fluid [CSF], dura etc., which are included in references to "tissue”), that is not predominantly bony tissue.
  • a CNS target tissue site may be located on or within or in proximity to (collectively, "on") the spinal cord, h one embodiment, acoustic scatter data is acquired at a target site that is on the middle cerebral artery (MCA) using TCD techniques, as described below.
  • MCA middle cerebral artery
  • acoustic properties of, such as acoustic scatter from, cranial blood vessels and other physiological structures that traverse the brain or are in communication with CNS tissue sites, such as the carotid and vertebral arteries may be taken at locations outside the cranial cavity, and provide good target sites for measurement of acoustic data in methods of the present invention.
  • exemplary methods and systems of the present invention are described with reference to acquisition of acoustic data from target sites on the MCA, it will be recognized that target sites on other cranial blood vessels that are in communication with or traverse the CNS may also be used.
  • Ultrasound detection techniques are preferred for assessing the acoustic properties of target CNS tissue sites for many embodiments.
  • Ultrasound sources and detectors may be employed in a transmission mode, or in a variety of reflection or scatter modes, including modes that examine the transference of pressure waves into shear waves, and vice versa.
  • Detection techniques involving measurement of values for or changes in acoustic scatter, such as back scatter or forward scatter, or reflection, and particularly backscatter are preferred for u se i n m any e mbodiments o f m ethods a nd sy stems o f t he p resent i nvention.
  • Exemplary acoustic data that may be used to determine ICP according to the present invention include: values for or changes in acoustic scatter, including values of and changes in the amplitude, phase and/or frequency of acoustic signals, values for or changes in length of scattered signals relative to the interrogation signal, values for or changes in the primary and/or other maxima and/or minima amplitudes of an acoustic signal within a cardiac and/or respiratory cycle; values for or changes in ratios of the maximum and/or minimum amplitude to that of the mean or variance or distribution of subsequent signals within a cardiac cycle, values for or changes in temporal or spatial variance of scattered or emitted signals at different times in the same target location and/or at the same time in different target locations, values for or changes in endogenous and/or induced brain tissue displacement or relaxation, and rates of change for such displacements, such as the velocity or acceleration of displacement, and the like, and combinations of these data.
  • acoustic interrogation signals may be employed, at the same or different frequencies, pulse lengths, pulse repetition frequencies, intensities, and the multiple interrogation signals may be emitted from the same location, or multiple locations, simultaneously and/or sequentially.
  • Acoustic scatter data may be collected, for example, from a cranial blood vessel at different points along the vessel, within or outside the cranial cavity, or from multiple sites at or in proximity to different vessels, or from multiple CNS tissue sites. Scatter from single or multiple interrogation signals may be detected at single or at multiple frequencies, at single or multiple time points, and at single or multiple locations.
  • methods and systems of the present invention may be used to localize differences in ICP within CNS tissue, thereby localizing areas of trauma or dysfunction. This may be achieved by acquiring acoustic data from a plurality of CNS sites and processing the plurality of data sets to determine ICP at corresponding multiple spatial locations within CNS tissue.
  • a property of intracranial blood flow may be determined in any blood vessel that traverses, or enters or exits CNS tissue (collectively, "cranial blood vessels"), with arteries being preferred, and the Middle Cerebral Artery (MCA), carotid and vertebral arteries being especially suitable.
  • Intracranial blood flow properties may be determined using any noninvasive or minimally invasive modality and is preferably determined using ultrasound techniques such as Transcranial Doppler (TCD) ultrasound techniques, which are well known in the art.
  • TCD Transcranial Doppler
  • TCD techniques are used to measure flow velocity in the MCA (V_mca), and V_mca measurements are used alone, or with other physiological parameters such as ABP, and/or CNS tissue displacement, to determine ICP.
  • V_mca flow velocity in the MCA
  • cranial blood vessel and/or blood flow characteristics, and other CNS tissue properties may alternatively or additionally be measured, or predicted, using other modalities such as noninvasive optical detection techniques, such as near infrared spectroscopic (NTRS) techniques.
  • NTRS near infrared spectroscopic
  • two variable inputs namely cranial blood flow velocity (or acoustic scatter data collected from a target site on a cranial blood vessel) and arterial blood pressure (ABP)
  • U ltrasound t echniques are p referably u sed, n on-invasively, t o measure an acoustic property of cranial blood vessel(s) or blood flow, such as flow velocity.
  • TCD is a preferred ultrasound technique and can provide substantially continuous measurement of flow velocity.
  • TCD devices are known in the art and may be used for collecting the acoustic back scatter and/or cranial blood flow velocity data used as a variable input for ICP determinations of the present invention.
  • the Spencer Technologies TCD 100M Power M-Mode Digital Transcranial Doppler device is one such suitable device.
  • ABP is preferably measured using noninvasive techniques, although invasive techniques may be used. Cranial blood vessel and/or blood flow characteristics and ABP may be measured on a substantially continuous or an intermittent basis. Blood flow velocity measurements may be determined on a substantially continuous basis, or on an intermittent basis, using TCD techniques.
  • ABP may be determined on a substantially continuous basis using, for example, an i nvasive a rterial 1 ine.
  • ABP may also be measured non-invasively, on an intermittent or substantially c ontinuous b asis using acoustic techniques as described in PCT International Publication No. WO 02/43564.
  • ABP measurements may also be measured non-invasively using, for example, the VASOTRAC ® device manufactured by Medwave, Inc., 4382 Round Lake Road West, St. Paul, MN 55112-3923.
  • ICP is determined by measuring an acoustic property relating to a cranial blood vessel, such as acoustic backscatter produced when an ultrasound beam is focused on a cranial blood vessel, or blood flow velocity measured, for example, using TCD techniques, and then using a non-linear relationship to relate the acoustic property and/or blood flow velocity to ICP.
  • the non-linear relationships used to predict ICP in methods and systems of the present invention may be derived empirically, using first principles, or using a combination of empirical data with first principles.
  • the non-linear relationship between blood vessel and/or blood flow properties may be derived, for example, based on non-linear empirical analytical methods such as the use of Hidden Markov Models, Support Vector Machines, Artificial Neural Networks, cellular automata and non-linear filters, as well as non-linear numerical methods.
  • non-linear empirical analytical methods such as the use of Hidden Markov Models, Support Vector Machines, Artificial Neural Networks, cellular automata and non-linear filters, as well as non-linear numerical methods.
  • ICP may be determined by measuring an acoustic property relating to a cranial blood vessel, such as acoustic scatter produced when an ultrasound beam is focused on a cranial blood vessel, or blood flow velocity, and then using a linear relationship between b lood v essel and/or b lood flow o r C NS t issue p roperties (as d escribed above) t o relate the acoustic property and/or blood flow velocity to ICP.
  • Linear relationships may be derived, for example, using first principles methods based on linear differential equations or based on linearized equations derived from non-linear fluid dynamics equations such as Navier-Stokes e quations.
  • a neural network set up and trained as described below, was used to derive a nonlinear relationship, which is further characterized below, and which provided accurate, determinations of ICP in experimental protocols based on two variable parameters: Vjmca measurements taken using TCD techniques; and ABP measured using an arterial line.
  • Vjmca measurements taken using TCD techniques
  • ABP measured using an arterial line The ABP data collected using an invasive arterial line was also computationally adjusted to mimic ABP data that could be collected non-invasively using, for example, a pressure cuff. Accurate ICP determinations using V_mca measurements and the adjusted ABP data were also demonstrated using methodologies and systems of the present invention.
  • Data such as V_mca data and ABP data, may be acquired intermittently or on a substantially continuous basis. Data is preferably collected at least twice in a cardiac cycle. By measurement on a “substantially continuous basis” we mean collection at least four data points, and preferably at least six data points per cardiac cycle.
  • substantially continuous basis we mean collection at least four data points, and preferably at least six data points per cardiac cycle.
  • methods and systems of the present invention may be used to non-invasively determine the autoregulation status of a patient together with, or separately from, a determination o f ICP, ABP, and other CNS properties.
  • C hallenges resulting in a modulation of the arterial blood pressure may be administered, for example, by having a subject perform actions that modulate the ABP in a predictable fashion, by adjusting intra- thoracic pressure using a ventilator, by restricting blood flow to an extremity, or by administering an agent, such as a diuretic and/or vasodilator or vasoconstrictor, that modulates arterial blood flow, may be used with methods and systems of the present invention to assess autoregulation.
  • Methods and systems of the present invention are preferably integrated in a controller component having data processing, storage and display features that provide meaningful information to professional clinicians.
  • the controller component may be integrated with other clinical devices, or may be programmed to receive additional data inputs relating to other clinical parameters.
  • a "long term" ICP trace corresponding to ICP determinations made over a time period of at least several minutes and up to several hours or days is provided to illustrate trends and fluctuations in ICP determinations over time. ICP determinations taken over time and relating to a particular patient may also be stored and displayed, in a variety of formats, to illustrate ICP trends over various time periods.
  • a "short term" I CP trace showing substantially instantaneous I CP measurements taken over two or more cardiac cycles may also be provided.
  • An exemplary data display unit is described below.
  • methods and systems of the present invention provide spatial location of desired target areas based on their acoustic properties and automated focusing of an acoustic source at the desired target area. Suitable source/detector combinations and transducer assemblies for scanning and locating desired target areas are also described.
  • Fig. 1 shows exemplary output from the pulse boundary calculation used to delineate the beginning and end of each cardiac cycle.
  • the upper trace represents the ABP signal.
  • the lower trace marks systole (positive tick), and diastole (negative tick).
  • Fig. 2 shows an example of a training set for a given patient using an ANN as described below.
  • Each input vector record has 42 values and is made up of an ABP pulse, V_mca Pulse, instantaneous heart rate, and static pressure difference, concatenated together.
  • Fig. 3 shows an exemplary signal record produced by patient manipulation during data acquisition.
  • the ABP record (the upper trace) shows evidence that this patient has undergone a blood draw from the arterial line.
  • the invasive ICP trace (lower trace) is not affected.
  • Fig. 4 shows the predicted ICP (lower trace) compared to measured ICP (upper trace) based on the simple mathematical first principles model described below.
  • Fig. 5 illustrates the major cerebral vessels, including the middle cerebral artery (MCA), the target of standard transcranial Doppler procedures, and schematically illustrates an acoustic source emitting acoustic interrogation signals in a scanning mode.
  • MCA middle cerebral artery
  • Figs. 6A an 6B show, schematically, the use o f a transducer array o f the present invention in a scanning mode (Fig 6A) used to locate the target area of interest based on its acoustic properties, and in a focusing and data acquisition mode (Fig. 6B).
  • Fig. 7 shows a schematic illustration of a single cMUT transducer cell structure.
  • Fig. 8 shows a schematic illustration of a cMUT transducer array comprising a plurality of cMUT transducer cell structures.
  • Fig. 9 shows a schematic diagram of an acoustic source/detector combination comprising a combination PVDF/cMUT tranducer structure.
  • Figs. 10A and 10B show schematic diagrams of an acoustic source/detector combination comprising combination PVDF array/PZT transducer structure.
  • Figs. 11A and 11B illustrate an exemplary patient interface unit having an acoustic source/detector combination of the present invention.
  • Figs. 12A and 12B show comparisons of instantaneous traces of invasively measured (generally, lower trace) and non-invasively determined I CP (generally, upper trace) for an exemplary patient described in Example 1.
  • Fig. 13 shows traces of invasively measured (upper trace at left) and non-invasively measured (lower trace at left) ICP, averaged with a moving box-car filter on a one-minute time scale.
  • Fig. 14 shows the mean values for invasively measured ICP plotted against mean values for non-invasively detemiined ICP with data collected over a 10 minute period and subjected to a one minute running average.
  • Fig. 15 shows the mean values for invasively measured ICP plotted against mean values for non-invasively determined ICP with data collected over a 10 minute period and subjected to a one minute running average.
  • Fig. 16 shows application of an algorithm formulated using a neural network and a 29 patient training set to determine ICP for each of the 29 patient members of the training set using acoustic scatter and ABP data.
  • Fig. 17 shows application of the algorithm formulated using the 29 patient training set described above to the acoustic backscatter and ABP data for each of 10 patients that were not part of the 29 patient training group.
  • Fig. 18 illustrates application of an algorithm formulated using a neural network and acoustic scatter and ABP data from a 25 patient training set to determine ICP for each of 21 validation patients that were not part of the 25 patient training set.
  • Fig. 19 illustrates application of an algorithm fo ⁇ nulated using a neural network and acoustic scatter and ABP data from the 25 patient training set in an efficacy demonstration to determine ICP for each of 6 patients that were not part of the training or validation patient sets.
  • One aspect of methods and systems of the present invention relates to determination of ICP using variable input data that can be measured using non-invasive and/or minimally invasive measurement techniques. Methods and systems for determining ICP based on noninvasive or minimally invasive measurements of V_mca and or ABP are described in detail. While the methods and systems of the present invention may be embodied in a variety of different forms, the specific embodiments shown in the figures and described herein are presented with the understanding that the present disclosure is to be considered exemplary of the principles of the invention, and is not intended to limit the invention to the illustrations and description provided herein.
  • variable parameters such as acoustic back scatter from within, on or in proximity to a cranial blood vessel alone, or in combination with ABP and/or other variable parameters and ICP.
  • Determination of ICP based on physiological measurements taken using non-invasive or minimally invasive techniques including ultrasound techniques such as measurement of acoustic scatter and/or Doppler flow velocity , e.g. in the middle cerebral artery (V_mca) and invasive or non-invasive (e.g. cuff or tonometric) arterial blood pressure (ABP), among others, may be accomplished using a variety of empirical methods.
  • V_mca is measured (non-invasively) using Doppler techniques
  • ABP is measured (non- invasively) in a cranial blood vessel such as the middle cerebral artery using "active" ultrasound techniques, as described herein.
  • ICP prediction may be implemented using linear filters, including those with both infinite impulse response (IIR) and finite impulse response (FIR) properties.
  • Linear filter operations are capable of scaling (i.e. filter output is proportional to input) and superposition (i.e. the sum of filter outputs from two separate inputs is the same as the filter output from the sum of two inputs). Such operations are described in Chen C, Linear System Theory and Design, Oxford University Press, 1999.
  • time-invariance i.e. an . operation having the same characteristic, such as frequency response, at all points in time
  • the properties of the input signal(s) could dynamically dictate the response characteristics of the linear filter portion of the algorithm.
  • ICP may be represented as a sum and/or the convolution product of one or more samples of ABP and V_mca, or other TCD-derived measurement(s) such as a tapped delay line in which two or more samples of data, usually with fixed sample interval, are considered simultaneously, with an impulse response vector.
  • the impulse response vector is adjusted through well-known means (e.g., least-squares error minimization) so that the error produced by the operation is minimized when applied to the data gathered from real patients.
  • Certain properties of the cardiovascular system are demonstrably non-linear. That is, they do not meet the linear systems criteria of scalability and superposition described by Hashizume, Y., 1988, "Non-linear Pressure Wave Propagation in Arteries," Journal of the Physical Society of Japan, Vol.57, N12, pp. 4160-4168. For this reason, standard linear systems approaches such as frequency decomposition may not work sufficiently well in an analytical approach.
  • These non-linear properties include the variation of the viscosity of blood with respect to shear rate and the visoelasticity of arterial walls, among others.
  • Non-linear methodologies may be used, under some circumstances, with linear methodologies to provide non-invasive ICP determinations. ICP determinations maybe derived, for example, using a non-linear, empirically derived, relationship in combination with a linear first principles relationship.
  • a non-linear operation called a non-linear filter can be used to describe the relationship between ICP and related physiologic measurements including ABP and V_mca.
  • a non-linear filter is any operation for which the properties of scaling and superposition do not hold true. Therefore, each input to the non-linear filter has a unique output that cannot be described by the standard methods of linear systems theory.
  • an empirical model may be constructed using non-linear filter theory.
  • Empirical relationships between the variable inputs discussed herein and the desired determination, such as ICP may be derived using numerous mathematical techniques, including correlations, artificial neural networks, non-linear regression methodologies, Baysian statistical methods, artificial life methodologies, and the like. Exemplary techniques are described below.
  • HMM Hidden Markov Model
  • a system may be in one a finite set of states, hidden from the outside world, but which can be inferred from emissions, or observable phenomena, which are particular to each state.
  • HMM Hidden Markov Model
  • Such models are described, for example, in Boyer, X., Koller, D, Tractable Inference for Complex Stochastic Processes, Technical report, Stanford University, U S A, 1998.
  • the Bayesian probability of transition from one state to any other may be inferred from observing the emissions of a system over time. All relevant information about a system described by a HMM can be inferred from the current state of a system, rather than prior states the system may have adopted.
  • ICP can be thought of as taking on a finite number of unknown states
  • state 1 equals 1 mmHg
  • state 2 equals 2 mmHg
  • minimally invasive or noninvasive measurements such as ABP or V_mca may be considered as emissions from which the state of the system (ICP value) can be inferred.
  • This HMM system could be trained to calculate the probability of a transition from one ICP level to another based on experimental measurements. These probabilities can then be used to calculate the likelihood that the system is at a particular state, or ICP level, based on recent ABP and Vjmca measurements.
  • the following rule may be based on observations regarding the relationship between ICP, ABP, and V_mca: if arterial pulse pressure ⁇ equals> 'within normal limits' and systolic BP ⁇ equals> 'within normal limits' and diastolic BP ⁇ greater than> 'within normal limits' and V_mca ⁇ equals> 'significantly reduced flow' and vaso spasm criteria ⁇ equals> 'not met' then ICP ⁇ equals> 'elevated' Dozens or hundreds of such rules may be constructed based on empirical data or on the output of first-principles mathematical models. This approach may be used as the primary means to diagnose elevated ICP or as an adjunct method to classify results or predict the likelihood that specific pathologies
  • ANN Artificial Neural Network
  • Neural network analysis is a well- described and important signal-processing teclmique that has found a number of uses in the medical field, including voice recognition, radiologic image analysis, and physiologic signal processing.
  • An AJSTN is a mathematical construct inspired by the biological nervous system, in which connection weights between individual network units known as neurons determines the relationship between network input and output values.
  • ANNs can reproduce any continuous function with an arbitrary level of accuracy. Blum E, Leong K, Approximation theory and feedforward networks, Neural Networks 1991; 4:511-515.
  • ANNs have been shown to be of particular value in the prediction of time series data, as described by Eisner JB, Predicting time series using a neural network as a method of distinguishing chaos from noise, J. Phys.
  • ANNs are trained to simulate a given relationship between input and desired target or output values through exposure to a training set; that is, a set of data in which inputs are matched to known output or target values.
  • a training set that is, a set of data in which inputs are matched to known output or target values.
  • the error between known target values and actual ANN output is used in an iterative fashion to modify network connection weights so as to minimize network error.
  • a validation (or test) data set similar to the training set but not used in network training is used to validate network performance.
  • an ANN can be produced which is likely to perform well when exposed to real-world input data.
  • the training set would consist of inputs, containing data derived from ABP and V_mca measurements, and matched target values, containing data derived from invasive ICP measurements.
  • a network trained on a data set suitably representative of the clinical population on which the filter is to be used would be able to predict ICP from future ABP and V_mca input data.
  • Analytical methods may then be applied to input data from an unknown patient to determine how closely the input data of that patient matches each of the patient sets used to create the subset neural networks. The degree of match could then be used to determine how much each of these subset networks contributes to the final ICP prediction.
  • a methodology involving a particular neural network, represented by a unique combination of connection weights produced through training with a representative training set, may be fixed once it has demonstrated the ability to appropriately predict or simulate the invasive ICP recorded as part of a representative validation set not included in the training set.
  • the network used in clinical practice may be fixed and, if fixed, would not undergo any changes with exposure to patients it encountered in practice.
  • both experimental and commercial devices implementing a neural network to determine ICP may contain calibration network elements that are not fixed, but that facilitate individualized calibration of the device to a particular subject, or to a subset of subjects having certain characteristics, or to a subset of conditions having predetermined characteristics.
  • an important first step in data conditioning is the preparation of a monolithic, synchronized database file.
  • synchronized database file(s) for a given patient's data contains the data at a uniform rate.
  • a 250 Hz downsampled data rate is used because it contains nearly all important physiologic information and is industry- standard, in that it is commonly used by other investigators.
  • TCD and telemetry data records conventionally contain an external synchronization signal which has been fed to both devices during data acquisition. The data may be synchronized through cross-correlation analysis and alignment of the digitized external synchronization signal stored simultaneously in both the TCD and telemetry d ata records.
  • Synchronized database files such as 250Hz database files, produced from each patient's data forms the basis of all further data preparation and analysis.
  • the signals stored in the database are synchronized in an absolute sense, they remain out of phase with one another as a consequence of the physical locations n the body at which the measurements were generated.
  • the cardiac pulse does not reach the radial artery at the same time as it reaches the middle cerebral artery, so these signals will remain out of phase with one another even though they are synchronized with respect to time. This discrepancy is taken into account during building of the training set, described below.
  • the particular form in which data is stored is arbitrary, although the experimental database format was designed to allow data to be easily imported into Matlab, a commercial computing environment designed for algorithm development.
  • ANN software may be prepared and almost a limitless array of possible network topologies which might be selected for use in the methods and systems described herein.
  • ANN software and network topologies other than those described below may be used in methods and systems of the present invention and are know to those having ordinary skill in the art.
  • a non-linear multilayer perception (a 2-layer feed-forward ANN) was chosen as the experimental prototype architecture for its well-known characteristics and relatively straightforward training process.
  • This system is described in the following publications: Eisner JB, Predicting time series using a neural network as a method of distinguishing chaos from noise, J. Phys. A: Math 1992 25:843; M ⁇ ller MF, Efficient training of feed-forward neural networks, PhD thesis, Computer Science Department, Arhus university 1993; Riedmiller M.
  • the hidden neuron layer utilizes an approximation of the hyperbolic tangent function as a transfer function, which allows the network to model non-linear input- target relationships; the output neuron layer uses a linear transfer function so that network
  • outputs can be linearly scaled.
  • Network inputs consist of normalized, arbitrary-duration tapped delay-lines of invasive ABP and Doppler ultrasound V_mca data in which each input contains pulse-contour data from one or more cardiac cycles.
  • Network output represents a continuous ICP pulse contour normalized to the duration of one cardiac cycle.
  • Network input, hidden layer, and output size is somewhat arbitrary and the best configuration for a given problem must often finally be determined through trial-and-error.
  • the system of the present invention employs an ANN with an input size of 42 samples (20 ABP pulse-contour samples, 20 Vjtnca pulse-contour samples, 1 instantaneous heart-rate value, and 1 value representing the measured static pressure difference between the invasive arterial line and the middle cerebral artery), a hidden layer consisting of 5 neurons, and an output layer consisting of 20 ICP pulse-contour samples.
  • Alternative input data formats include non-invasive blood pressure (derived through cuff, tonometric, or other means) which is r epresented as systolic, mean, and diastolic pressure values that are updated intermittently.
  • RNNs Recurrent Neural Networks
  • Elman supra
  • Giles C Lawrence S
  • Tsoi A noisy time-series prediction using a recurrent neural network and grammatical inference, Machine Learning 2001 July/Aug; 44(1):161-183.
  • This family of topologies maintains an internal memory of previous network inputs and/or outputs for improved time-series prediction.
  • RNNs can maintain an infinite impulse response.
  • RNNs present training challenges, as described in Atiya A, Parlos A, New results on recurrent network training: unifying the algorithms and accelerating convergence, IEEE Trans. On Neural Networks 2000 May; 11(3), but they are promising and would be suitable for use in methods and systems of the present invention.
  • Neural network software designed as part of this research program was implemented in Matlab 6.5 using portions of the Mathworks Neural Network Toolbox, a commercial neural network computing package used extensively in industry.
  • the format of the data p resented to the ANN is somewhat arbitrary, provided that data from multiple sources is synchronized, hi one embodiment, synchronized 250 Hz data undergoes transformation from the time-domain to the pulse-domain and is presented to the network as a series of pulse-contours represented by a fixed, arbitrary number of data points normalized to the duration of one or more cardiac cycles.
  • Many other implementations are suitable, and several have been tried with varying results.
  • the pulse-domain implementation presented here has been the most effective of the methods tried. It is important to understand that the data presented to the network is essentially absolute-valued pressure and flow data (following a linear scaling), and can therefore be used to track and predict absolute ICP values.
  • pulse-domain trans fo ⁇ nation and input/target set building occurs is straightforward but somewhat involved.
  • the process is generally done in a piecemeal fashion so that a manageable amount of data is handled at any one time (e.g. 30 seconds of data may be processed at a time).
  • the steps are as follows:
  • Phase synchronization As mentioned earlier, the database record is synchronized with regard to absolute acquisition time, but contains cardiac-cycle phase discrepancies between signal records.
  • the first step in training set preparation is to align ABP, V_mca, and ICP records (e.g. with cross-correlation spectrum analysis and alignment) so that they are in phase with regard to cardiac-cycle boundaries. This results in multiple records, one for each physiologic signal, that are in phase with one another.
  • FIG. 1 shows exemplary output from the pulse boundary calculation used to delineate the beginning and end of each cardiac cycle, with the upper trace representing the ABP signal and the lower trace marking systole (positive-going) and diastole (negative- going).
  • each linear signal record from each subsequent cardiac cycle is isolated and resampled using standard signal-processing techniques such that it occupies a fixed, arbitrary number of data points.
  • Experimental work was carried out with a pulse-domain width of 20 samples, although other pulse domain widths would be operative and suitable.
  • each signal can be thought of as occupying a two-dimensional array in which each successive column contains successive pulse beats normalized to a fixed pulse-domain width.
  • Moving window Doppler envelope Although successive resampled ABP pulse beats derived in this way generally represent smooth continuous data that is appropriate for presentation to a neural network, V nca flow data as it is currently gathered contains noise and is best represented as a pulse-envelope over a moving window of time. In other words, each point in pulse-time (a particular sample index in a given normalized pulse) can be considered to be best represented by the maximum Doppler velocity occurring i n a window of arbitrary width o f s uccessive n ormalized p ulse b eats.
  • Neural networks operate best when input data falls within a relatively narrow range (e.g. -1 to 1), so the data at this stage must be normalized between defined extremes (e.g. physiologic values likely to be encountered in clinical use). Because these normalization extremes are fixed, the normalization operation represents a reversible transformation. This means that although the input values have been rescaled, they still represent absolute values.
  • defined extremes e.g. physiologic values likely to be encountered in clinical use.
  • Each input vector preferably c ontains elements of each signal r ecord (ABP and V_mca), so the normalized pulse-domain elements from step 5 are concatenated to one another.
  • instantaneous heart rate and static pressure head measured between the site of blood pressure measurement and the MCA is concatenated to each input vector.
  • the data may be concatenated within or across multiple cardiac cycles, or may be normalized to other cyclical (or non-cyclical) physiological events. These values may not be necessary, but they may improve network performance.
  • the input records are now complete. (7) Preparation of matched target records - For training to be successful, each input vector in the training set must have a matched target vector.
  • the current implementation defines the target vector as one or more cardiac cycles of ICP data resampled to a fixed number of values.
  • the ICP data is therefore normalized in time, but since it is synchronous with the input data, the ICP pulse can be stretched or shrunk using the original heart rate data stored in the input set so that it represents a realistic ICP pulse contour.
  • the target set is built at the same time as the input set, and undergoes each of the steps above except for step 6, since target vectors do not currently contain any additional information beside ICP.
  • Exemplary data from a training set for a patient is illustrated in Fig. 2. There are approximately 1500 individual cardiac cycle records displayed pulse by pulse. Each input vector record has 42 values and is made up of an
  • Pulse domain input and/or target data c an be stored to disk or created in real-time from telemetry and Doppler flow data. Patients representative of those likely to be encoimtered in clinical practice (by virtue of ICP, gender, ethnicity, pathology, etc.) are selected to populate the training panel. Each of these patients' 250 Hz database(s) (or the equivalent linear records stored in memory) undergoes training set preparation and is included i n t he s et o f i nput-target v ectors t hat a re p resented t o t he n eural n etwork d uring training.
  • ANNs are trained through exposure to a well-defined data set representative oft hat 1 ikely t o b e encountered i n r eal- world p ractice.
  • error between network output and known target output is passed to a training algorithm, which adjusts network connection weights in order to minimize e ⁇ or.
  • network e ⁇ or is minimized.
  • n etwork training methodology Proper selection of the n etwork training methodology is important, because it may influence both the efficiency and accuracy of ANN training.
  • the training e ⁇ or should usually be the minimum achievable level of e ⁇ or. This is a valid assumption as long as the network is not "over-powered" (possessing more hidden nodes than necessary) and will not over- fit the data.
  • An excellent discussion of this problem is given by Caruana et al. in Caruana R, Lawrence S, Giles C, Overfitting in neural nets: backpropagation, conjugate gradients, and early stopping, Neural Information Processing Systems, Denver CO 2000 Nov 28-30.
  • a number of different networks with varying numbers of hidden nodes and varying degrees of training must be used and the results of the validation input set compared to the known validation ICP target data.
  • networks with 5, 10, 15, 20, and 25 hidden nodes may be trained for 500, 1000, 1500, 2000, 2500 and 3000 epochs.
  • Analysis of the validation set e ⁇ or produced by each network delineates which architecture is the suitable for the given training and validation s et.
  • T hreshold v alues for the v alidation s et e ⁇ or i.e. acceptable 1 evels o f ICP e ⁇ or
  • An additional aim of the experimental process was to identify subgroups of patients which may share common noninvasive ICP measurement characteristics (e.g. on the basis of vascular properties) and which may be identified on the basis of ABP or TCD characteristics (e.g. hypertension, vasospasm), mechanism of injury, physical exam or laboratory findings, patient demographics (e.g. age, sex, weight), or a combination of these parameters.
  • noninvasive ICP measurement characteristics e.g. on the basis of vascular properties
  • ABP or TCD characteristics e.g. hypertension, vasospasm
  • mechanism of injury e.g. hypertension, vasospasm
  • physical exam or laboratory findings e.g. age, sex, weight
  • patient demographics e.g. age, sex, weight
  • SOM Kohonen Self-Organizing Map
  • This methodology uses an unsupervised type of neural network closely associated with biological memory in which differing network inputs are mapped to unique regions of a dimensionally-reduced output space.
  • unknown patients' data would map to regions of the SOM representing input characteristics most similar to a particular patient or subgroup of the input set, and the network trained on that data would be chosen to non-invasively predict ICP.
  • These methods are expected to improve noninvasive ICP determination in specific patients or subgroups of patients with vascular parameters differing appreciably from the majority of patients in the training set.
  • the methods described above for neural network training preferably result in one or more fixed methodologies for predicting ICP based on individual patient input and variables.
  • Systems and methodologies of the present invention comprehend both the experimental systems and methodologies used to derive relationships between variable parameters such as V_mca and/or ABP and ICP, such as neural networks that undergo training and validation, as well as clinically useful and commercial methodologies and systems implementing one or more fixed methodologies for determining ICP based in individual patient data resulting from neural network training and validation.
  • Validation of the network performance is achieved by presenting validation or test data from one or more patients (patient data identical to training set data in that invasive ICP and ABP are known, but which has not been included in the training set) to the ANN and comparing the subsequent predicted ICP output to the test patients' known or target invasive
  • the neural network employed as part of the research described above may be configured to be perfectly stable, in that its connection weights are fixed and are not influenced by the type or order of data presented to its inputs. The response characteristics of the final methodology must be carefully delineated.
  • a second stage of this characterization is through network validation, described above.
  • a second stage is through the relatively straightforward step of input space mapping, in which thousands of canonical inputs encompassing the input set likely to be encountered in clinical practice (e.g. ABP waveforms throughout the physiologic range in increments of 1 mm Hg, each with V_mca wavefomis tliroughout the physiologic range in increments of 1 cm/s) are presented to the validated network, and the resultant ICP outputs recorded.
  • This process characterizes the network response over the entire input space likely to be encountered.
  • Such input-mapping highlights any problem areas in terms of input space response, and checks the physiologic ranges over which the network can be expected to perform adequately.
  • Figure 3 shows an example of a signal record produced by patient manipulation during data acquisition.
  • the ABP record (upper trace) shows evidence that this patient has undergone a blood draw from the arterial line.
  • the invasive ICP trace (lower trace) is not affected. Most disruptions that occur during data acquisition are similarly obvious, lending themselves to automated detection.
  • the patient's ABP and ICP data collection devices may be recalibrated prior to and monitored continuously during data acquisition, with the time and duration of any manipulation during data acquisition (e.g. blood draw) being recorded.
  • the database record entries co ⁇ esponding to these recorded times of patient manipulation may be marked for exclusion from training set data.
  • the synchronized (e.g. 250 Hz) database records may then be inspected (e.g.
  • V_mca, ABP, and ICP curve abnormalities e.g., significant blunting suggesting arterial line malposition and V__mca dropout from a malpositioned data acquisition device. Any abnormal records are marked for exclusion from training set data. Signal values that do not remain in the expected physiological ranges are also excluded.
  • Manual data checking and/or automated software methods may be implemented. The data-checking function is preferably implemented as a fully automated software process.
  • ICP data output derived using the methodology described above can be characterized in a number of ways to ensure that relevant physiological values are conveyed to the user. These include methods to assure that the range and rate of change of the ICP wavefo ⁇ n is physiologically reasonable (e.g. noninvasive ICP should be a positive value less than 75 mmHg), that the ICP pulse shape, height, and other parameters are realistic, and that there is a reasonable degree of concordance between the forcing functions inherent in the system (ABP and V_mca) and the system output (ICP).
  • Neural networks can be considered to operate using Bayesian probability estimates, and can therefore also provide a measure of confidence in a given ICP prediction, a capability that has been utilized in other medical signal processing tasks, as described in Dorffher G, Can neural networks improve signal processing? A critical assessment from the ANDEE project, NEC Research Institute Online Archive - http://citeseer.nj.nec.com/186775.html. Providing high and/or low confidence values allows the instmment and/or the health care provider to give particular weight to the predicted ICP value or to discard it, or to prompt the provider to make sure that all sensors and lines are properly attached to the patient.
  • ABP and V_mca data are preferably acquired and processed in an integrated electronic device and may therefore be conveniently synchronous with respect to acquisition time, eliminating the need for data synchronization.
  • ABP and V_mca may be acquired using different devices and/or synchronization rates, with the data being collected and processed in an integrated processing unit that provides data synchronization, as necessary.
  • (2) Downsampling / resampling of telemetry and Doppler flow data This allows each linear signal record to occupy the same amount of space so that standard signal processing techniques may be employed more easily.
  • Pulse domain transformation Transfoimation of the linear, phase-aligned, time- domain telemetry and Doppler flow records to two-dimensional, normalized pulse-domain records may be desirable. This is a multi-step process and may involve calculation and storage of beat-to-beat instantaneous heart rate, normalization of each cardiac cycle to a fixed number of samples, and moving pulse-window smoothing or envelope calculation for the V_mca Doppler flow data.
  • the raw data output from the trained and validated network is a pulse-domain record normalized to lie on the interval
  • ICP preferably provide trend analysis and data display features.
  • One suitable output display provides: (1) one or more trace(s) of ICP over a "long term" period of time of at lest several minutes and up to several hours or days to illustrate trends in patient ICP; (2) a trace of "instantaneous" or “short term” ICP, determined over several cardiac cycles; and (3) additional graphical representations that may aid in guidance of an acoustic transducer or transducer a ⁇ ay, as described below.
  • a representation of flow velocity vs. depth of the transducer focus may be provided.
  • a display showing both instantaneous ICP determinations over fewer than ten (10) previous cardiac cycles and ICP determinations over a period of at least several minutes is preferred. It will be evident to one of ordinary skill in the art that data may be displayed in a variety of ways.
  • ANN Using an ANN, as described above, is a convenient and reliable teclmique for deriving an accurate non-linear relationship between patient input variables, such as V_mca and ABP and the detemiined output, ICP.
  • patient input variables such as V_mca and ABP
  • ICP the detemiined output
  • CAs are mathematical constructs in which a regular, discreet lattice of cells, each of which can adopt one or more of a finite set or continuous range of states, evolve over successive discrete time steps according to specific rules regarding each cell's state and the state(s) of that cell's neighbor(s).
  • CA's and similar modeling techniques the propagation of natural phenomena through time can be modeled in reproducible ways that do not require extremely complex mathematical description. Exemplary techniques are described in Smith MA, Cellular Automata Methods in Mathematical Physics, PhD Thesis, Massachusetts Institute of Technology, May 1994.
  • the propagation of the CA model may then be used to produce statistical or expert systems describing the predicted behavior of the physical system after which the CA has been modeled.
  • ICP prediction for instance, a simple one-dimensional system of cells whose internal states describe blood flow or ABP and/or another physiological property, may be used to model the propagation of the cardiac pulse through the cerebral vasculature over time.
  • the propagation rule may take into account physical factors such as the elastance of the arterial wall, the viscosity of blood, central venous pressure, the level of ICP, and other properties that could influence blood flow.
  • Detennination of ICP based on related physiological parameters measured using noninvasive or minimally invasive teclmiques may also be approached using first-principles methods based on non-linear relationships, or on linear differential equations.
  • Such a system may be represented in the form of a more commonly encountered engineering system, such as an electrical transmission line, to which appropriate teclmiques from electrical engineering analysis may be applied and a closed form solution obtained, as described in Ursino M, Lodi CA, Interaction among autoregulation, CO2 reactivity, and intracranial pressure: a mathematical model, Am J Physiol. 1998 May;274(5 Pt 2):H1715-28.
  • linearized equations in which non-linear terms are neglected or modified to simplify the solution, may be derived from non-linear fluid dynamics equations (e.g. Navier-Stokes equations), and the closed form solution obtained or solved numerically, as described in Olufsen MS, A one-dimensional fluid dynamic model of the systemic arteries, Stud Health Technol Inform. 2000;71:79-97.
  • ICP is an important variable in any reasonably inclusive system of derived equations describing cerebrovascular fluid mechanics, and may be solved for given ABP, Vjmca, and/or other physiologic data.
  • An exemplary linear first-order differential equation model of the cerebral vasculature was developed using invasive ABP, invasive ICP, and V_mca signals.
  • the radial artery ABP signal was taken as a surrogate for the ABP at the inlet of the MCA.
  • Flow is presumed to be primarily resistive.
  • Jugular venous pressure at the cranial outlet (JVP) is taken as the true outlet pressure of the cranial vasculature and is assumed to be zero.
  • JVP cranial venous pressure at the cranial outlet
  • ICP is assumed to act according to the 3 rd pressure principle, such that ICP acts in lieu of JVP as cranial outlet pressure to define the pressure drop between the cranial inlet and outlet.
  • Each patient is assumed to have a characteristic vascular resistance, k, that is unknown initially but that can be calculated from physiologic data, hi particular, the heart is treated as a step-function generator with a given pulse height and resultant flow impulse response. The relationship between these two quantities determines k. Volumetric flow is assumed to be directly proportional to V_mca. Finally, ICP is calculated as the required cranial outlet pressure given MCA inlet pressure, characteristic resistance k, and volumetric flow derived from V_mca.
  • This simple exemplary method utilizes intrinsic blood pressure and flow manipulations generated by the heart to dete ⁇ nine the characteristics of the cerebral vasculature, which are then used to predict ICP.
  • Fig. 4 shows the predicted ICP (lower trace at left axis) compared to invasively measured ICP (upper trace at left axis) based on the simple model described above that captures the step-down in arterial blood pressure as it enters the brain, with resistive, viscous flow in the middle cerebral artery, and the heart modeled as a step-function generator such that changes in arterial blood pressure lead to changes in cerebral blood flow.
  • This particular model was driven using the systolic and diastolic values of arterial blood pressure and flow in the middle cerebral artery to predict ICP.
  • first principles methods such as those described above are used in combination with an empirical approach, such as a neural network approach, to make ICP determinations non-invasively using acoustic scatter and/or ABP data.
  • First principles methods may be applied to patient data in the first instance to make a preliminary ICP determination, with empirical methods used on all or a portion of patient data to make co ⁇ ections to or adjust or refine the preliminary ICP findings.
  • the relationship between ICP, ABP, V_mca, and/or other physiological measurements may alternatively or additionally be modeled using well-known methods (e.g. finite element analysis) that allow the numerical solution of discretized versions of the Navier-Stokes equation, which describes the dynamics of fluid mechanical systems such as the c erebral v asculature.
  • Solution of the Navier-Stokes equation allows one to take many of the non-linear properties of the cardiovascular system into account, including nonlinear arterial viscoelasticity, convective momentum and, potentially, non-Newtonian viscosity.
  • First principles methods may be applied to patient data in the first instance to make a preliminary ICP dete ⁇ nination, with empirical methods used on all or a portion of patient data to rn al e c onections t o o r a djust o r r efine the p reliminary ICP findings, a s d escribed above.
  • Other combinations of empirical and first principles approaches may also be used, and exemplary methods are described below.
  • One method uses a derived relationship between spontaneous (intrinsic) tissue displacement (resulting from blood flow, CSF, etc.), detemiined by analyzing acoustic scatter from a CNS target tissue site, ABP, and invasively monitored ICP to determine ICP based upon invasively or non-invasively measured tissue displacement and ABP.
  • V_mca may be used instead of, or in combination with, measured tissue displacement as variables using a combined empirical/first principles approach.
  • using an ultrasound probe operating above 100 kHz a given volume of tissue is insonated with a wavefo ⁇ n having a specific frequency and amplitude, and the time or phase shift of a reflected ultrasound signal is used to calculate intrinsic tissue displacements.
  • d t*1500 m/sec
  • d tissue displacement
  • t the time or phase shift of the reflected signal
  • 1500 m/sec is the estimated speed of sound through the brain.
  • ICP CPP - MAP
  • MAP (2*diastolic ABP + systolic ABP)/3
  • d F(CPP)
  • F can be any function, such as an exponential, vector, matrix, integral, etc., or a simply an empirical relationship with CPP
  • CPP - MAP - ICP F 2 (d)
  • F 2 F 1 .
  • This method uses a derived relationship between the amplitude of reflected acoustic signal(s) from CNS target tissue sites, ABP, and invasively monitored ICP to estimate ICP from non-invasively measured acoustic signals and ABP.
  • a given volume of tissue is insonated with a waveform having a specific frequency and amplitude, and the amplitude of the backscatter is used to create a waveform of tissue reflection/absorption.
  • This new waveform, ⁇ can be generated by integrating the amplitude of the backscatter over a finite epoch (such as the cardiac cycle, measured with ECG tracing) and normalizing this by the time period of the epoch.
  • ⁇ can be normalized to the MAP (as defined above), to produce a waveform ⁇ .
  • the relationship between this normalized waveform, ⁇ , and invasively measured ICP is then determined by talcing simultaneous measurements of the backscatter signal, ABP, and ICP and solving for the equation
  • ICP F( ⁇ ), where F is any mathematical function, or simply an empirical relationship.
  • methods and systems of the present invention use existing methods for determining ICP, based on standard TCD measurements, replacing noninvasive measurements of V_mca with noninvasive measurements of the displacement of CNS tissue caused by blood flow, the cardiac cycle and respiration, or by combining one or more variables, including V_mca, tissue displacement, and other physiological variables.
  • TCD transcranial Doppler
  • the processing steps of the present invention utilize simultaneous and continuous measurements of invasive ICP, invasive or noninvasive ABP, and displacement (or the like) to generate a set of equations that accurately predict ICP using only non- invasively-detemiined displacement and ABP data alone.
  • Flow velocity and tissue displacement may replace the ABP measurement.
  • Step 1 A weight function is calculated between ABP and ICP, using a system of linear equations. The solution of this system of equations results in a vector containing the coefficients of the weight function. Any number of coefficients can be chosen to model this system. For example, we will select 25 coefficients.
  • Step 2 The coefficients of a weight function between displacement and ABP curves are used as movement characteristics. The computation is similar to the one described in Step 1 and perfo ⁇ ned at the same time. Again, any number of coefficients can be used; we will select 6 for this example.
  • Step 3 The relationships between the movement characteristics of Step 2 and the 25 coefficients of the weight function in Step 1 are described by an approximating linear function (i.e., matrix A and vector B), which is calculated through a sequence of 25 multiple regression analyses of the patients' data.
  • matrix A and vector B an approximating linear function
  • Steps 1-3 are perfo ⁇ ned, the noninvasive ICP determination is made as follows: while the displacement (or the like) and ABP curves are recorded non-invasively for a new patient (one not used in the derivation of the above simulation function), the movement characteristics are computed every 10 seconds and transfe ⁇ ed to the simulation function. Finally, the simulation function transforms the ABP curve into a simulated ICP curve.
  • Noninvasive systems and methods of the present invention provide a measure of arterial or venous blood pressure using acoustic techniques to measure alternating compression and dilation of the cross-section or other geometric or material properties of an artery or vein, using empirically established relationships and/or mathematical models, hi another aspect, blood pressure is determined using acoustic teclmiques to measure alternating compression and dilation of tissue su ⁇ ounding blood vessels that is displaced as the vessels are compressed and dilated with the cardiac c ycle.
  • Geometrical properties that may be detennined using acoustic detection techniques include changes in diameter, cross-sectional area, aspect ratio, rates of changes in diameter, velocity, and the like.
  • Material properties that may be determined using acoustic detection techniques include the stiffness of vessel walls or tissue in proximity to vessel walls.
  • Blood pressure may be assessed, for example, by acquiring acoustic data, in an active and/or passive mode, from target tissue sites at or in proximity to one or more blood vessels.
  • the acoustic data can be related to the stiffness of vessel walls or supporting tissue, which can be related to blood pressure, just as acoustic data from a CNS target tissue site can be related to tissue stiffness, which can be related to ICP.
  • Suitable target tissue sites for determination of arterial or venous blood pressure may comprise any blood vessel or su ⁇ ounding tissue. Detection of ultrasound scatter data may be related, for example, with synchronous Doppler flow measurements within the same vessel.
  • a calibration step using a measure of blood pressure taken with a conventional blood pressure device may be incorporated in the blood pressure determination.
  • Acoustic proxies for the pulsatility of the blood vessel - such as oscillation rate of the blood vessel wall - may be substituted for direct measures of those quantities, h this method, the spontaneous changes in the diameter (or other geometric property) of the vessel being monitored are assessed using ultrasovmd, and this information is related (e.g., using co ⁇ elation techniques) to synchronous Doppler flow measurements within the same vessel.
  • blood pressure can be calculated from flow velocity measured by Doppler. By simultaneously measuring the pulsatility of the blood vessel of interest and the Doppler flow velocity proximal and distal to this site, continuous blood pressure can be determined.
  • a patient's autoregulation status, or autoregulation capacity may also be determined using acoustic data related to intrinsic and/or induced tissue displacements according to the present invention, as described in greater detail below.
  • ICP and autoregulation status, or autoregulation capacity are intimately related.
  • the net volume of blood within the brain at any time point within the cardiac cycle is a function of systemic blood pressure and protective autoregulatory mechanisms of the brain vasculature, from its major arteries, having diameters on the order of millimeters, to its arterioles, having diameters on the order of microns.
  • These various physical scales of cerebral vasculature respond with different time scales and different levels o f c ontribution t o the d etermination o f ICP and autoregulation.
  • the different classes of cerebral vasculature have different material properties, such as Young's modulus, which contribute to the different displacement properties in the brain.
  • CPP cerebral perfusion pressure
  • the brain and its v asculature are capable of altering CPP in order to maintain proper blood flow to the brain. This is refe ⁇ ed to as a normal state of autoregulation.
  • autoregulation is abnormal and ICP becomes directly proportional to the mean arterial blood pressure.
  • CPP is determined from the displacement or emission data and ABP data.
  • co ⁇ elation coefficient indices between time averaged mean flow velocity (FVm) and CPP (Mx), and between the flow velocity during systole and CPP (Sx), are calculated during several minute epochs and averaged for each investigation. These co ⁇ elation indices are detemiined for a variety of clinical situations in which autoregulation and outcome is known.
  • regression lines are determined to infer the status of cerebral autoregulation for any set of Mx and Sx values. See, Czosnyka et al, Monitoring of Cerebral Autoregulation in Head-Injured Patients, Stroke Vol. 27, No. 10, October, 1996) hi another embodiment, continuously acquired noninvasive acoustic data relating to tissue displacement(s) and/or emission(s) is used along with simultaneous measurements of continuous ABP, to determine the status of cerebral autoregulation. Specifically, a pressure reactivity index (PRx) is calculated as a moving co ⁇ elation coefficient between a finite number of consecutive samples of values for displacement and/or emission and ABP averaged over several minutes.
  • PRx pressure reactivity index
  • a continuous index of cerebrovascular reactivity (autoregulation) to changes in ABP is determined.
  • a positive PRx is indicative of impaired autoregulation and predicts unfavorable outcome, while a negative PRx indicates intact autoregulation and likely good outcome. See, Czosnyka et al., Continuous Monitoring of Cerebrovascular Pressure-Reactivity in Head Injury, Acta Neurochir [Suppl] 71:74-77, 1998).
  • spectral analysis of simultaneously acquired continuous, noninvasive acoustic data relating to tissue displacement(s) and/or emission(s) and continuous invasive or noninvasive ABP data is used to determine the status of cerebrovascular autoregulation.
  • Transfer functions (TFn) are calculated from fast Fourier transform (FFT) spectra as ratios of displacement and/or emission and ABP harmonic peak amplitudes to distinguish states of vasoreactivity.
  • FFT fast Fourier transform
  • TF are calculated for a variety of known clinical conditions, and this data is used to determine values for the TF that co ⁇ espond to specific states of autoregulation.
  • TF values can differentiate impaired autoregulation from e ffects s olely related to e levated ICP o r a ctive v asodilation.
  • the hemodynamic and/or cerebrospinal systems may need to be perturbed for a finite period of time to cause a known alteration in ICP, or to challenge autoregulation.
  • perturbations involving physiological challenges, are described below:
  • Mechanical perturbations of the hemodynamic system for evaluation of autoregulation may involve the placement of large pneumatic or hydraulic blood pressure cuffs around the lower extremities and inflated in order to increase venous return to the heart, thereby increasing vascular blood volume, leading to increased blood flow to the brain.
  • the state of autoregulation can be assessed by analysis of the Doppler information.
  • Other means of increasing blood flow to the brain including placing the patient in a gravity suit, changing ventilatory parameters on mechanical ventilators for intubated patients, and restricting arterial blood flow to the periphery.
  • Pharmacological perturbations of hemodynamic system for evaluation of autoregulation If autoregulation is intact, the brain can respond to this decreased blood flow by re-directing blood flow and altering resistance to ensure that it receives adequate perfusion.
  • intravenous fluid boluses can be administered to transiently increase blood volume and flow to the brain. If autoregulation is intact, the brain can respond accordingly.
  • Other means for altering the blood volume and flow include the use of vasopressors, vasodilators, chronotropic and contractility agents.
  • Trendelenberg position and changes in patient equilibrium, such as coughing, sneezing, etc., that alter ICP.
  • Acoustic source/detector assemblies scanning and localization methodologies
  • One aspect of the present invention relates to acoustic source/detector assemblies for use in methods and systems of the present invention.
  • an acoustic source/detector combination such as a TCD source/detector, is stably mounted, or held, in proximity to a patient's body surface, such that the focus of the acoustic source(s) is adjustable to provide an acoustic focal point on a blood vessel or other target site within the patient's body.
  • the acoustic source/detector is stably mounted, or held, in proximity to a cranial window, such that the focus of the acoustic source(s) is adjustable to provide an acoustic focal point on CNS tissue, such as on a cranial blood vessel.
  • the acoustic source/detector combination is preferably provided as a unitary component, but separate acoustic source and detector components may also be used.
  • the acoustic source/detector combination may be provided in connection with a mounting structure or accessory that provides temporary adherence to desired patient sampling locations and may be provided as a single use component.
  • acoustic transducers and acoustic transducer a ⁇ ays may be used as acoustic source/detector assemblies and acoustic data acquisition components of the present invention.
  • a single acoustic transducer, or a singer acoustic transducer a ⁇ ay may be operated both as a source and a detector, or separate source and detector transducers or transducer a ⁇ ays may be provided.
  • Conventional PZT acoustic transducers may be implemented as acoustic data acquisition components in methods and systems of the present invention.
  • Acoustic transducer a ⁇ ays composed of cMUT and PVDF cells or elements may also be used and are prefe ⁇ ed for many implementations.
  • PZT, cMUT and PVDF acoustic transducers and a ⁇ ays may be combined in various data acquisition components and operated in acoustic source and/or receiver modes in yet other embodiments.
  • the acoustic source/detector combination may be mounted on a stabilizer, or on or in a structure, such as a helmet-type structure or headband that may be mounted on the head.
  • An applicator containing an acoustically transmissive material such as an acoustic gel, may be placed between the surface of the acoustic source/detector combination and the head. Steering of the acoustic device may be accomplished manually or using automated mechanisms, such as mechanical or electronic steering mechanisms. Such mechanisms are well known in the art.
  • One drawback of using acoustic techniques for measuring physiological parameters using a standard TCD transducer is that localization of a desired CNS target area using an acoustic transducer is challenging and often requires a trained, experienced sonographer to find and (acoustically) illuminate the desired target area, such as the MCA.
  • the sonographer After locating the desired target area, the sonographer generally attaches a cumbersome and uncomfortable headset to the transducer that stabilizes the transducer position and reduces the effects of patient movement and other disturbances on the position of the transducer. The sonographer may also be required to monitor acoustic readings and reposition the transducer intermittently to maintain the focus on the desired data acquisition area.
  • Fig. 5 illustrates the major cerebral vessels, including the middle cerebral artery (MCA) 10, the target of standard transcranial Doppler procedures and a target for acoustic measurements used in the methodology employed for detennining ICP described above.
  • MCA middle cerebral artery
  • the anterior cerebral arteries 14, anterior connecting artery 16, internal carotid artery 18 and posterior connecting artery 19 are shown.
  • the darkened blood vessel branches denote blood flow towards acoustic device 12, while cross-hatched blood vessel sections denote flow away from the transducer.
  • An acoustic source detector assembly 12 of the present invention is illustrated to the right of the cerebral vessels, emitting a coustic interrogation signals in a scanning mode as d escribed below, in which a large target area is acoustically illuminated prior to the localization of a smaller target site.
  • another aspect of the present invention relates to methods and systems for locating and acoustically illuminating and/or probing a desired target site in an automated fashion using an a ⁇ ay comprising a plurality of acoustic source and/or detector elements.
  • An acoustic transducer/receiver a ⁇ ay may be employed in a scanning mode, for example, to acquire acoustic data from numerous sites within a larger target area. Based on the acoustic data collected in the scanning mode, localized sites within the target area may be selected as target sites for focused acoustic illumination and/or probing.
  • Localized target sites may be selected, or predetermined, based on any aspect of the acoustic data collected in the scanning mode, such as acoustic scatter amplitude, phase and/or frequency maxima or minima, tissue stiffness properties, endogenous and/or induced tissue displacement properties, rates of change of such properties, and the like. Focusing of elements of the acoustic transducer/receiver a ⁇ ay on selected target sites may be accomplished in an automated fashion using mechanical or electronic beam steering and other automated acoustic focusing methodologies.
  • an automated system locates a desired target site within a larger target area in a scanning mode, focuses on the desired target site for acquisition of acoustic data, and thereafter periodically scans the target area and repositions the acoustic focus, if necessary, to maintain the focus of the acoustic source at the desired target site.
  • Multiple target sites may also be located in a scanning mode and focused on sequentially and/or simultaneously for acoustic data acquisition from multiple target sites using acoustic transducer/receiver array assemblies of the present invention.
  • Systems incorporating suitable a ⁇ ays of acoustic source and/or detector elements are disclosed.
  • Fig. 6A illustrates, schematically, the use of a scanning acoustic transducer assembly 20 of the present invention that acoustically illuminates and acquires acoustic data from multiple points within a broad target area 22, such as a large portion of the cerebral blood vessel complex, in a scanning mode. Based on the acoustic data acquired in the scanning mode, localized target sites 24 within the scanned area may be identified and elements of the transducer assembly are focused on localized target site(s) for acquisition of acoustic data from the desired target site(s), as shown in Fig. 6B.
  • Selection of localized target site(s) may be predete ⁇ nined based on various acoustic properties, including the amplitude (or any amplitude derivative) of acoustic scatter data, Doppler analysis of acoustic scatter data, phase or frequency of acoustic data, changes in the primary and/or other maxima and/or minima amplitude, phase or frequency of acoustic signals within a cardiac and/or respiratory cycle or other period, or determinations derived from acoustic data, such as flow velocity, tissue stiffness properties, endogenous and/or induced tissue displacement properties, acoustic emissions associated with such displacements, rates o f change o f s uch p roperties, and the like.
  • acoustic properties including the amplitude (or any amplitude derivative) of acoustic scatter data, Doppler analysis of acoustic scatter data, phase or frequency of acoustic data, changes in the primary and/or other maxima and
  • the selection of a desired localized target site is preferably accomplished by scanning the desired target area, as shown in Fig. 6 A, and determining the localized site of highest amplitude acoustic scatter, or highest Doppler or flow velocity values, which represents the MCA. Acoustic elements of the acoustic source/receiver data acquisition component may then be focused on one or more localized MCA sites for acoustic data acquisition.
  • NIRS near infra-red spectroscopy
  • magnetic resonance and other techniques are known and used, for example, to image and locate internal physiological structures.
  • Such techniques may be used in association with the methods and systems of the present invention for locating internal physiological structures prior to assessment of acoustic properties.
  • an acoustic source/detector combination preferably an acoustic transducer a ⁇ ay comprising multiple transducer elements, is operable in both a scanning mode and a focusing mode.
  • One or more acoustic source element(s) of the acoustic data acquisition component scan(s) target areas of the interior of the cranium (or another target area) in a scanning mode to identify target sites having predete ⁇ nined or desired acoustic properties.
  • one or more of the acoustic source(s) may be manually or automatically focused on the desired target site(s) for operation in an acoustic inte ⁇ ogation or data acquisition mode.
  • the acoustic source may also be programmed to monitor acquired acoustic data and to adjust the positioning and/or focus of the source to maintain the focus of selected or predete ⁇ nined acoustic source(s) on the desired target site.
  • the acoustic source(s) may be programmed to collect data from a plurality of predete ⁇ nined or programmed target sites at predetermined time points.
  • Acoustic transducer source and detector elements of the present invention may, in fact, be programmed to collect one or more types of acoustic data from a single or multiple target sites, at one or more times. Acquisition of acoustic data, using methods and systems of the present invention, is preferably accomplished in an automated fashion.
  • Range-Doppler processing is an efficient implementation of matched filtering that has been used in the radar i and sonar signal processing community for many years. It is a robust technique, in part because it makes very few assumptions about the statistical nature of the environment and targets that it encounters. Range-Doppler processing provides a useful decomposition of the spatial and temporal (i.e. Doppler) scattering properties of the target of interest.
  • Sensor time series data are divided into frames, often overlapped, multiplied by the transmitted waveform replica and then transfomied into the frequency domain via the Fast Fourier Transform (FFT) algorithm.
  • FFT Fast Fourier Transform
  • These operations implement, very efficiently, a bank of matched filters, each matched to a narrow range of Doppler shifts.
  • Range-Doppler processing affords separation of targets in terms of their range and speed relative to the acoustic device.
  • MCA flow is by far the largest target, which makes it a natural for this 'search and home in' approach.
  • signals from a target are combined in a convolution with signals from a reference source after each is measured on an acoustic a ⁇ ay.
  • the net result is a formula whose maximum occurs at the target site.
  • all of the acoustic fields may be replaced by the Fourier transform of the acoustic field, or a component of the Fourier transform of the acoustic field, e.g.
  • the Doppler signal h this embodiment, the Fourier transform of the acoustic backscatter from an acoustic a ⁇ ay serves as the target signal, and the forward scatter from a TCD or a ⁇ ay placed on the opposite temple may be used as the reference source.
  • These signals would be mathematically combined to find and maintain an acoustic focus on a desired target area.
  • the feature of interest will be known to represent a local if not global minimum or maximum among a spatial distribution of values of the feature of interest.
  • An exemplary acoustic system providing an automated targeting feature while allowing user participation in targeting may utilize conventional TCD systems made by DWL, Spencer Technologies, Nicolet, etc., where the acoustic sensor consists of a single transducer element, and where the acoustic system provides information only along the beam of the single transducer for a given orientation of that transducer.
  • the user manually manipulates the transducer so that it insonifies different portions of the cerebral architecture, and electronically steers the depth along the transducer beam axis.
  • the user would be guided by the real-time display of information, along with the user's memory of what the display has shown in the preceding moments, to seek out the maximum in flow velocity in the MCA.
  • One portion of the display may provide the real time value of the variable of interest at a position relative to the face of the transducer (reported in absolute units, or arbitrary units, since the actual depth is not important) that is chosen by the user with a cursor designated for this purpose.
  • the display may provide, for example, the real time value of flow velocity in the MCA, otherwise known as the spectrogram of the flow.
  • Another portion of the display may provide a graphical image designed to communicate to the user, at any given orientation of the transducer, the direction of larger values of flow in the MCA relative to the real time position of the cursor. This may take the fo ⁇ n of two arrows pointing in different directions, e.g. one pointing 'up' one pointing 'down,' where up and down are known to the user to represent deeper relative to the present position of the cursor, and more shallow relative to the present position of the cursor, respectively. If there are local maxima in flow velocity in both directions, the direction in which a greater maximum exists would be designated by having a brighter a ⁇ ow pointing in that direction.
  • These flow velocity gradients may be calculated within the associated controller component by measming the Doppler shift along all of the points insonified at a given moment by the transducer to provide a real-time calculation of the local gradient of the flow velocity. This calculation may be performed using a variety of well-known mathematical formulae (one-sided differences, centered differences to a variety of orders, etc).
  • the absolute position of the local flow velocity maximum in flow in the MCA need not be known or reported or displayed to the user. What the user gains from this analysis is a direction, relative to the cu ⁇ ent position of the cursor, which position need not be defined, of the local maximum in flow velocity.
  • the user may then manipulate the cursor to report the spectrogram at a deeper or a shallower position along the acoustic beam and judge for themselves whether they have achieved a local maximum in flow velocity.
  • Standard TCD devices also allow for the device to emit sound whose amplitude is tied to the flow velocity at a given point along the beam of the transducer, the one, in particular, whose spectrogram is shown to the user.
  • supplemental infomiation would be of interest to the user of the present invention.
  • an acoustic a ⁇ ay comprising a relatively dense distribution of acoustic transducers rather than a single transducer or a sparse a ⁇ ay, one may have, at any given moment, infonnation relating to the relative spatial distribution of flow velocity in depth at a variety of angles from the center of the acoustic beam.
  • a user assist feature may provide a display showing the direction of the local flow velocity maximum.
  • locational infonnation relating to the direction of maximum flow velocity may be provided in additional dimensions, and the user may be guided by an arrow pointing in each of the three possible directions of cursor movement relative to the real time cursor position.
  • One set of a ⁇ ows may indicate the local maximum is deeper than, or shallower than, the present cursor position. Another set of a ⁇ ows may indicate that the local maximum is more anterior or posterior to the present cursor position. Yet another set of a ⁇ ows may indicate that the local maximum in flow velocity is more superior or inferior to the present cursor position. This information may be calculated as described above, using Doppler analysis of acoustic backscatter from the field of positions insonified by the transducer a ⁇ ay.
  • the user's positioning of the a ⁇ ay may be guided by this information, along with supplemental aural and visual information a s described above, including the instantaneous spectrogram at the position of interest, to move the cursor, and re-examine the spectrogram.
  • MCA middle cerebral artery
  • Acoustic transducer a ⁇ ays of the present invention are generally thin and generally comprise a single layer or thickness of transducer elements. Stacked, multiple layer transducer cells, or elements, may be used for some applications.
  • the transducer elements or cells may be a ⁇ anged on a single plane to form a generally flat, planar a ⁇ ay, or they may be a ⁇ anged to fonn a curved or a geometrically stepped a ⁇ ay.
  • Transducer anays having various configurations and structures may be useful for applications contemplated in this disclosure.
  • data acquisition components comprising acoustic source/detector combinations of the present invention comprise a plurality of capacitive micromachined ultrasound transducer (cMUT) cells.
  • cMUT ultrasound transducers are manufactured using semiconductor processing techniques and have sufficient power and sensitivity to transmit and receive at diagnostic ultrasound energy levels, which is necessary and sufficient for purposes of the present invention.
  • the transducer elements are fabricated using small capacitive diaphragm structures mounted on a silicon substrate.
  • cMUT transducer a ⁇ ays have the potential of being produced very inexpensively, and may also have the support electronics integrated onto the same chip.
  • Fig. 7 shows a schematic illustration of a single cMUT ultrasound transducer cell structure.
  • cMUT ultrasound transducer cell 40 comprises a positive electrode 42 illustrated as the top electrode and a negative electrode 44 illustrated as the bottom electrode.
  • the top electrode is generally provided on or in connection with a flexible membrane and the bottom electrode is generally provided on or in connection with a substrate 46, such as a silicon substrate.
  • Insulating supports 48 are provided to form a sealed chamber 50 between the positive and negative electrodes.
  • Chamber 50 may contain a gas or liquid or gel-like substance, or it may be provided as an evacuated chamber.
  • the diaphragm structures of the cMUT ultrasound transducer convert acoustic vibrations into a modulated capacitance signal, or vice versa.
  • cMUT transducer elements may be operated in various modes of transmit and receive operation, including unbiased mode, non-collapsed mode, collapsed mode and collapsed snapback mode (transmit only).
  • cMUT transducer cells, elements and a ⁇ ays One advantage of using cMUT transducer cells, elements and a ⁇ ays is that the electronics may be provided on or in the cell structure, greatly simplifying the electronic communication to and from the a ⁇ ay and facilitating programmable a ⁇ ay features.
  • a cMUT transducer a ⁇ ay is composed of multiple individual cMUT ultrasound transducer cell structures a ⁇ ayed as elements, with the elements anayed in rows and/or columns and/or smaller divisions forming the a ⁇ ay.
  • Fig. 8 schematically illustrates such an a ⁇ ay 60.
  • the number of cMUT transducer cells 62 forming each of the transducer elements 64, and the number of elements forming the a ⁇ ay may be varied, depending on the a ⁇ ay application.
  • cMUT transducer a ⁇ ays having various configurations may be assembled and used in the present invention.
  • cMUT transducer a ⁇ ays can be configured and operated to achieve acoustic transmission and sensitivity levels sufficient to perfo ⁇ n as acoustic transmit/receive devices suitable for use in medical devices, such as TCD devices. More specifically, cMUT transducer anays of the type illustrated in Fig.
  • cMUT transducer anays operated experimentally at an 80Vbias and at a gain of 60 and 80 dB to receive signals from CNS target sites in a range of less than 4 to greater than 6 cm from the anay at a level sufficient to make Doppler dete ⁇ ninations.
  • cMUT transducer cells and elements may be a ⁇ anged in different c ombinations to provide cMUT transducer a ⁇ ays having different capabilities. If each of the cMUT cells is provided with independently controlled or controllably electronics, each of the cMUT cells may act as a transducer element and an a ⁇ ay may be provided as a plurality of independently controlled or independently controllable cMUT cells. More typically, a transducer element comprises a plurality of cMUT cells that is electronically controlled or controllable as a unit. Thus, in the a ⁇ ay illustrated in Fig.
  • each of the elements 64 is composed of multiple (6 X 6) cMUT transducer cells 62 that are controlled or controllable as a unit.
  • a plurality of the elements 64 such as elements forming a row or a column, may be electronically controlled or controllable as a unit to provide a cMUT transducer a ⁇ ay comprising a plurality of row or column transducer elements.
  • a one-dimensional (ID) anay may be composed of a single transducer element comprising multiple cells, while a two- dimensional (2D) a ⁇ ay is composed of multiple transducer elements a ⁇ anged in a generally planar, two-dimensional configuration.
  • two cMUT acoustic a ⁇ ays are aligned in a "Mills Cross" configuration in winch two transducer a ⁇ ays are a ⁇ anged generally orthogonal to one another, which allows one anay to sweep vertically in send and receive modes and the other to sweep horizontally in receive and send modes, hi this implementation, a first linear cMUT transmit anay may be steerable in a first direction, such as a vertical direction and a second linear cMUT receive a ⁇ ay is a ⁇ anged generally o rthogonal t o the first linear a ⁇ ay and m ay b e s teerable in a direction orthogonal to the first direction.
  • the two, crossed linear cMUT a ⁇ ays alternatively transmit 5 and receive ultrasound beams while steering the sending and listening beams, to identify and focus on acoustic signals having the desired property.
  • an acoustic a ⁇ ay comprising PVDF (polyvinylidene fluoride) film transducers is used as an acoustic detector a ⁇ ay, alone or in combination with a cMUT a ⁇ ay or a single element PZT transducer employed as 'the source, an exemplary
  • PVDF anay in combination with another transducer or a ⁇ ay
  • the source transducer or anay transmits sound through the PVDF a ⁇ ay, sweeping the sound in a single dimension generally perpendicular to the anangement of the PVDF anay.
  • the PVDF a ⁇ ay serves as the acoustic detector, receiving and processing acoustic signals.
  • Fig. 9 presents a schematic diagram illustrating an acoustic transducer a ⁇ ay 70 of the
  • present invention comprising combination PVDF/cMUT a ⁇ ays.
  • the combined depth of the a ⁇ ays is generally quite small and may be on the order of about 1 cm.
  • the cMUT a ⁇ ay 72 is a ⁇ anged below the PVDF a ⁇ ay 74, with the PVDF a ⁇ ay 74 a ⁇ anged closest to the subject's surface during use. In this configuration, the cMUT a ⁇ ay operates as the acoustic source and transmits acoustic beams through the PVDF a ⁇ ay.
  • cMUT a ⁇ ay 72 may be composed of a
  • the PVDF a ⁇ ay may also be provided as a ID anay (as shown) or as a 2D a ⁇ ay.
  • acoustic source(s) and/or detector(s) are provided as 2D a ⁇ ays, they are capable of sending and/or detecting acoustic signals in two dimensions, rather than a single direction.
  • Figs. 1 0A and 1 0B show a s chematic diagram illustrating an acoustic anay o f the 5 present invention comprising combination PVDF a ⁇ ay/PZT transducer.
  • a cMUT a ⁇ ay may similarly be used in combination with a PZT transducer.
  • the PVT transducer is generally mounted below the PVDF or cMUT anay and transmits as an acoustic source through the PVDF or cMUT a ⁇ ay in a single, broad beam, hi these embodiments, the PZT transducer generally serves as the acoustic source and the PVDF or cMUT a ⁇ ay generally serves as the 0 acoustic detector.
  • Fig. 10A shows an acoustic source/detector combination 80 comprising a PZT transducer 82 underlying a PVDF or cMUT transducer anay 84 having a plurality of aligned transducer elements 86. Each of the aligned transducer elements 86 is controlled or controllable as a unit.
  • Fig. 10B illustrates another acoustic source/detector combination 90 5 comprising a PZT transducer 92 underlying a PVDF or cMUT transducer a ⁇ ay 94.
  • transducer a ⁇ ay 94 comprises a plurality of transducer elements 96 controlled or controllable as a unit spaced in a two-dimensional configuration.
  • the PVDF or cMUT a ⁇ ay may be constructed as a ID a ⁇ ay comprising a plurality of aligned transducer elements, as shown in Fig. 10A, or as a 2D a ⁇ ay comprising a plurality of transducer elements ananged in a two dimensional configuration, as shown in Fig. 10B.
  • One of the advantages of the ultrasound transducer anay systems of the present invention is that multifunctional arrays may be provided in a relatively high power, yet inexpensive system.
  • Such a ⁇ ays are very versatile, are capable of performing multiple acoustic functions and may be pre-programmed or programmable to provide desired functions, and may be provided as disposable or single-use elements of an integrated clinical diagnostic system.
  • acoustic a ⁇ ays of the present invention are provided as a single-use acoustic data acquisition component of a medical device, such as an ICP monitoring device, comprising one or more acoustic transducer a ⁇ ays in operative communication with a controller component having data processing, storage and/or display capability.
  • the one or more acoustic transducer a ⁇ ays may communicate with the controller component by means of one or more detachable cables, or using a radio frequency, infrared or other wireless technology.
  • the transducer anay(s) may be steerable and may be programmed to scan one or more target areas having certain boundaries or parameters, and locate one or more desired target site(s) based on preselected or selectable acoustic properties.
  • the transducer anay(s) may furthermore be programmed and/or controllable to establish and maintain a focus by directing ultrasound beams having a preselected intensity, amplitude, phase, frequency, etc., to the target site(s) in an automated fashion.
  • Transducer a ⁇ ays of the present invention may also be programmed to collect acoustic data from multiple target sites simultaneously, or at different times.
  • a transducer anay may be programmed to operate alternatively as acoustic sources and detectors.
  • multiple transducer a ⁇ ays used for monitoring multiple patients provide data to and communicate with a single data processing, storage and display device.
  • Figs. 11A and 11B illustrate one exemplary embodiment of acoustic data acquisition components comprising acoustic source/detector systems, such as acoustic anays, of the present invention
  • acoustic source/detector systems such as acoustic anays
  • both disposable and non-disposable elements are shown
  • costly elements of the acoustic system are provided as non-disposable components, while less costly components, which require close interaction with a patient and, perhaps, sterilization, are provided as a single-use component.
  • Fig. 11A illustrates an acoustic data acquisition component 100 comprising an acoustic transducer anay 102 that interfaces with an anay electronics component 104 and an acoustic transmission component 106 that facilitates high fidelity acoustic transmission between transducer a ⁇ ay 102 and a subject's body surface.
  • Acoustic transmission component 106 preferably comprises a sealed enclosure containing an acoustically transmissive material, such as an acoustic gel having uniform properties and being substantially free from acoustically significant discontinuities, such as bubbles.
  • Acoustic transmission component 106 may incorporate an adhesive substance on a least a portion of an exposed surface 108 to facilitate temporary adherence of the data acquisition component to a subject's body surface. Exposed surface 108 bearing an adhesive substance may be protected by a detachable cover 110 that may be removed prior to placement on a subject's body surface.
  • the transducer array and a ⁇ ay electronics component may be pe ⁇ nanently mounted in or on a structure 112 that facilitates communication of data and/or power to and/or from a controller component.
  • Structure 112 may incorporate control and/or power features or may provide operable connection of the transducer anay and a ⁇ ay electronics to control and/or power features that are housed in a separate controller component.
  • Data acquisition component 100 may communicate with a controller component through a structure 112 and cable 114, as illustrated in Fig. 11 A, or communication may be provided using alternative communications methodologies, such as RF communications systems.
  • acoustic transmission component 106 may be provided as a single use component and may be affixed to an exposed surface of transducer anay 102 prior to mounting on a subject's body surface.
  • acoustic transducer a ⁇ ay 102, a ⁇ ay electronics component 104 and acoustic transmission component 106 may be provided as a single use acoustic data acquisition component 116, as illustrated schematically in Fig. 1 IB.
  • Single use acoustic data acquisition component 116 has an electronics interface component, illustrated schematically as wire 118, that provides communication between anay 102 and a ⁇ ay electronics component 104 and electronics and/or power capabilities provided in structure 112 or in a remote controller component.
  • the electronics interface component provided in connection with data acquisition component 116 may be a hard- wired interface component that relies on contact with a mating interface component in stracture 112, or it may be provided as a wireless interface communications component, h this embodiment, single use data acquisition components 116 may be packaged in a sterile or non-sterile fashion.
  • an acoustic anay is provided as part of a single use or disposable system element, in combination with a patient interface component.
  • the acoustic anay is preferably i n c ontact w ith a coustically t ransmissive m aterial, s uch a s a n a coustic gel, t hat provides high fidelity acoustic transmission into and from the target area.
  • the acoustically transmissive material is preferably interfaced with a contact material, such as an adhesive material, that facilitates temporary positioning and affixation of the disposable system element to a patient's skin.
  • the patient contact material may be protected by a removable cover, which is removable at the time of use.
  • the disposable system element, including the acoustic array may be provided as a unitary element that may be sterilized and packaged for one-time use.
  • acoustically transmissive material layers may be provided as a separately sterilized, packaged component that is designed to interface with a non-disposable component including the acoustic a ⁇ ay(s). Such layers may be provided with an adhesive layer on one side for contact with the patient's skin. Or, a recess may be provided for manual application of acoustically transmissive material. It will be evident that many different embodiments and a ⁇ angements of disposable and non-disposable elements may be employed.
  • This compact, disposable anay element may be placed in contact with the temple of the patient and, when activated, electronically scans a target area of interest, such as the area of cerebral blood vessels, and then focuses the acoustic source(s) and detector(s) on the target site o f interest, such as the M CA. T he acoustic a nay monitors and stays focused on the target area of interest during operation.
  • the acoustic a ⁇ ay forms part of a disposable assembly including an acoustic gel, or another acoustic material that facilitates transmission of acoustic signals at the interface with the patient's skin during operation.
  • the exposed surface of the acoustic gel is preferably interfaced with one or more adhesive elements that facilitate temporary placement on and consistent contact with a desired patient surface.
  • a removable cover may be provided over the acoustic gel to preserve the acoustic a ⁇ ay and other components.
  • Non-disposable elements of the system may include mounting hardware, one or more cables or wireless transmission interfaces, and a data processing, storage and display device (not shown). Placement of the acoustic source(s) and detector(s) on a subject for assessment of acoustic properties of CNS tissue (including blood and blood vessels) may be at known "acoustic windows" in the cranium.
  • the placement of the source(s) with respect to the detector(s) will depend on the acoustic data desired - e.g., for collection of back scatter acoustic data, the source(s) and detector(s) are in proximity to one another, while the source(s) and detector(s) are positioned generally opposite one another for collection of forward scatter acoustic data. Acoustic scatter or reflection data may be collected at various angles by placing the source(s) and detector(s) at various locations on the patient.
  • a hydrostatic sensor may be provided when the acoustic source/detector device comprises an acoustic, microwave or infra-red receiver, with a co ⁇ esponding sender being provided on the peripheral blood- pressure monitor.
  • the acoustic source/detector device comprises an acoustic, microwave or infra-red receiver, with a co ⁇ esponding sender being provided on the peripheral blood- pressure monitor.
  • a sensor that measures the direction towards the ground may also be placed on the acoustic source/detector device to create a coordinate system that allows the automatic measurement of the angle between the ground and the peripheral blood-pressure monitor.
  • ICP determinations using V_mca measurements although acoustic properties from other cranial target sites may be used in determining ICP. Automated acoustic scanning and target location may be facilitated using displayed information.
  • the user may simply mount a transducer a ⁇ ay on the patient, and an automated acoustic source/detector scanning feature operates to find desired targets, such as maximum V_mca. Other sites having unique acoustic properties may also be located. Coordinates for target locations and values for acoustic properties may be stored, over time, and displayed in a variety of formats.
  • Methods and systems of the present invention may be used in a variety of settings, including emergency medicine settings such as ambulances, emergency rooms, intensive care units, and the like, surgical settings, in-patient and out-patient care settings, residences, airplanes, trains, ships, public places, and the like.
  • emergency medicine settings such as ambulances, emergency rooms, intensive care units, and the like
  • surgical settings in-patient and out-patient care settings, residences, airplanes, trains, ships, public places, and the like.
  • the methods and systems of the present invention do not require patient participation, and patients that are incapacitated may also take advantage of these systems.
  • the methods and systems for assessing tissue properties, including ICP may be used on a continuous or intermittent basis for monitoring tissue properties or ICP.
  • a prototype system for collecting data, deriving and applying a non-linear relationship between the variables of cranial blood vessel velocity and ABP was assembled using commercially available components.
  • This prototype consisted of a notebook computer with a PCMCIA National Instruments (NI) 6024-E data acquisition (DAQ) card, a box containing the exposed backplane of the NI-DAQ card and a microphone input matching circuit, a specialized adapter designed to mate to the signal output port of a Spacelabs telemetry unit, and a Spencer Technologies TCD 100M Power M-Mode Digital Transcranial Doppler device and control pad with standard TCD ultrasound transducer and FDA-approved headband device for mechanical fixation to the head.
  • NI National Instruments
  • DAQ data acquisition
  • the Spencer TCD 100M device was not m odified i n a ny way from i ts F DA- appro ved c onfiguration.
  • UPS uninterruptible power supply
  • V_mca flow velocity in the middle cerebral artery
  • ABP arterial blood pressure
  • Figures 12A and 12B show a comparison of the instantaneous time traces of measured and predicted ICP from one of the core eight patients, reduced down from an initial data acquisition rate of 250 Hz to 20 Hz (i.e., there are twenty data points per second for each data set, or roughly twenty points per cardiac cycle).
  • the invasively measured ICP is shown in the generally lower traces, while the non-invasively infened ICP is shown in the generally upper traces.
  • the data in Fig. 12A manifests both cardiac and respiratory forcing, while the enlarged trace of Fig. 12B highlights the signals on the time scale of the cardiac cycle.
  • the predicted ICP traces are remarkably similar to the measured ones, with a tendency to under- predict the low values of ICP during diastole and over-predict the high values of ICP during systole. This is typical of seven out of our eight predictions, with an exception discussed below.
  • Point-by-point comparisons of time series for each patient demonstrates our successful prediction of invasively measured ICP, using only invasively measured ABP and acoustic measurement of the blood flow rate in the middle cerebral artery (V_mca), and a properly detemiined algorithm. Comparisons of invasively measured instantaneous ICP and predicted instantaneous ICP for eight patients are shown below, in Table 1.
  • the Invasive ICP column shows the mean and standard deviation of the invasively measured ICP.
  • the Predicted ICP c olumn s hows t he m ean a nd s tandard d eviation o f t he p redicted ICP .
  • T he E ⁇ or column shows the mean and standard deviation of the enor derived by subtracting the value of the measured and predicted ICP at each time point.
  • the comparison of invasive and predicted ICP for Patient #4 shows that the average point-by-point difference between invasive and predicted ICP was 0.80 mmHg, with the likelihood that 95% of the predicted ICP values will be within 1.68 mmHg of the measured value.
  • Figs. 12A and 12B show representative data from this representative patient. For six of the eight patients (#1-5 and 8), the average value of the instantaneous difference between measured and predicted ICP is less than 1 mmHg - the average uncertainty in the invasive ICP measurements. For another patient (#6), the average value of the instantaneous difference between measured and predicted ICP is less than 2 mmHg. Moreover, for these seven patients, the vast majority of the instantaneous differences are less than 2 mmHg.
  • the methodology does not need to predict instantaneous ICP. Short-term mean ICP values, and/or ICP trends, are sufficient for patient management. Therefore, the predicted ICP values were averaged and compared with averages of the invasively measured ICP.
  • time traces of invasively measured ICP (upper trace at left) compared well with time traced of predicted ICP (lower trace at left), as shown in Fig. 13.
  • a point-by-point comparison of these time series shows that their values lie within 1 mmHg of one another, which is equivalent to the documented uncertainty in invasively measured ICP. In practical terms, to achieve this level of accuracy, the system would have to process one minute worth of data before making a prediction of ICP, which is a clinically acceptable situation.
  • the mean and variance of the differences between invasively measured and predicted ICP were examined as a function of different averaging lengths. As the averaging length increases, this difference, along with the variance, decreases.
  • Our experimental observation was that a twenty second running average yielded a maximum variance that is within the documented uncertainty of the invasive ICP measurements and would therefore provide adequate reliability.
  • the practical implication is that the first predicted ICP value will be available twenty seconds following initiation of collection and processing of the input data. After that, the system output is a predicted running mean of ICP with a time scale of twenty seconds. This is a clinically useful output.
  • Example 1 The prototype device described in Example 1 and the nlCP determination methodology described in this specification was successfully tested on eighteen (18) patients at H arborview M edical Center i n S eattle W ashington, u sing b lood p ressure d erived e ither directly from an arterial line or using arterial line-based ABP data simplified to mimic ABP data obtained from a blood-pressure cuff.
  • H arborview M edical Center i n S eattle W ashington u sing b lood p ressure d erived e ither directly from an arterial line or using arterial line-based ABP data simplified to mimic ABP data obtained from a blood-pressure cuff.
  • a detailed description of the results for eight (8) of the eighteen patients was presented above in Example 1. Additional results are summarized below.
  • Fig. 15 shows the results of this analysis and demonstrates the feasibility of determining ICP based on variables measured using non-invasive techniques and implementing novel analysis of acoustic backscatter from the brain collected, for example, in a trans-temporal approach, and arterial blood-pressure data captured initially using an arterial line and then decimated for determining and testing the methodology in a way to mimic what one could leam about arterial blood pressure from a standard arterial blood pressure cuff that was used every 100 seconds.
  • the mean of actual, invasively measured ICP is plotted against the mean of predicted ICP determined over a ten-minute period, and subjected to a one- minute running average.
  • the variance shown on the figure is that of the difference between actual, invasively measured ICP and predicted ICP calculated after application of a one- minute running average.
  • training set of 29 patients, some of whom were included in the earlier 18 patient study.
  • ICP was invasively measured from the same hemisphere as the injury focus, and acoustic backscatter was measured from that side as well.
  • a neural network was trained as described herein using the 29 patient training set and an algorithm for dete ⁇ nining nlCP
  • the algorithm formulated using the 29 patient training set with the neural network was then applied to the acoustic backscatter and ABP data for each of the 29 patients in the training set to determine nlCP, and the non-invasively determined ICP was plotted against the invasively measured ICP.
  • the results are shown in Fig. 16.
  • the variance shown in Fig. 16 is that of the difference between (invasively measured) ICP and (non-invasively determined) nlCP calculated after application of a one-minute running average.
  • the (noninvasive) ICP determmation algorithm was highly effective in determining ICP non- invasively u sing t he a coustic b ackscatter a nd A BP d ata o n t he i ndividual m embers i n t he training set over a broad range of ICP values of from less than 10 mm Hg to nearly 30 mm Hg.
  • the algorithm fonnulated using the 29 patient training set as described above was then applied to the acoustic backscatter and ABP data for each of 10 patients that were not part of the 29 patient training group. Data was collected over a five to 20-minute period (depending on the patient) and subjected to a one-minute running average. The results are shown in Fig. 17. The variance shown is the figure is that of the difference between ICP and nlCP calculated after application of a one-minute running average. The results demonstrate that the algorithm formulated using the 29 patient training set was effective in determining ICP non-invasively using the acoustic backscatter and ABP data on new patients over a broad range of ICP values.
  • Example 1 Additional feasibility and efficacy testing of the methodology described above was performed using the experimental system described in Example 1.
  • Acoustic backscatter, ABP and invasively measured ICP data was collected from a set of 25 patients (the 'training set').
  • the ICP was invasively measured from the same hemisphere as the injury focus and acoustic backscatter was measured from this hemisphere as well.
  • the acoustic backscatter data was collected from the MCA and MCA flow velocity values were derived from the acoustic backscatter using conventional Doppler techniques. An empirical algorithm was derived using this data and the neural network training protocol described above.
  • the derived algorithm was then tested in an iterative fashion to determine ICP for 21 patients using only acoustic backscatter and ABP data for the 21 validation patients for whom invasively measured ICP data had also been collected.
  • the results of this validation testing are illustrated in Fig. 18, which plots the non-invasively determined ICP against the invasively measured ICP for each of the 21 validation patients.
  • the model algorithm provided a highly accurate detennination of ICP using only the acoustic backscatter and ABP data for fifteen of the 21 patients, shown as open circle data points in Fig. 18. ICP determinations for six of the 21 validation patients fell outside the predetermined acceptability standards, although four of the six outlying data points may be within an acceptable e ⁇ or range. Using a larger patient population for derivation of the ICP algorithm is anticipated to eliminate substantially all outlying patient ICP determinations.
  • the model algorithm derived during the validation testing was tested for efficacy and shown to be efficacious in a 6 patient sample for whom invasively measured ICP data had also been collected.
  • the non-invasively detemiined ICP is shown plotted against the invasively measured ICP for the 6 efficacy patients in Fig. 19.
  • Data from four of the six patients produced ICP determinations within or very close to the predetermined acceptability standards; the two other patients would be within an e ⁇ or range that may be acceptable for some purposes.
  • Using a larger patient population for derivation and validation of the ICP algorithm is anticipated to eliminate substantially all outlying patient ICP determinations.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Neurology (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Artificial Intelligence (AREA)
  • Neurosurgery (AREA)
  • Physiology (AREA)
  • Hematology (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Psychiatry (AREA)
  • Signal Processing (AREA)
  • Gynecology & Obstetrics (AREA)
  • Evolutionary Computation (AREA)
  • Psychology (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)
  • Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)

Abstract

L'invention concerne des systèmes et des procédés permettant de déterminer la PIC sur la base de paramètres pouvant être mesurés à l'aide de techniques non invasives ou avec effraction minimale, une relation non linéaire étant utilisée pour déterminer la PIC sur la base d'une ou plusieurs variables entrées. La première variable entrée correspond à une ou plusieurs propriétés d'un vaisseau sanguin crânien et/ou d'un débit sanguin, telles qu'une rétrodiffusion acoustique provenant d'un transducteur acoustique orienté vers un vaisseau sanguin crânien, la vitesse d'écoulement dans un vaisseau sanguin crânien, et analogue. D'autres variables, telles que la pression sanguine artérielle (PSA), peuvent être utilisées conjointement avec une première variable entrée correspondant à une ou plusieurs propriétés d'un vaisseau sanguin crânien, telles que la vitesse d'écoulement de l'artère cérébrale moyenne (ACM), pour dériver la PIC à l'aide d'une relation non linéaire. L'invention concerne également des procédés et des systèmes permettant de localiser des zones cibles sur la base de leurs propriétés acoustiques, d'effectuer un balayage acoustique d'une zone, d'identifier une zone cible d'intérêt sur la base de propriétés acoustiques et de concentrer de façon automatique une source acoustique et/ou un capteur sur une zone cible désirée. L'invention concerne également des ensembles de transducteurs acoustiques.
EP04754563A 2003-06-03 2004-06-03 Systemes et procedes permettant de determiner la pression intracranienne de fa on non invasive et ensembles de transducteurs acoustiques destines a etre utilises dans ces systemes Ceased EP1633234A4 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP11177783A EP2392262A1 (fr) 2003-06-03 2004-06-03 Procédés et systèmes pour localiser et illuminer acoustiquement une zone cible souhaitée

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US47580303P 2003-06-03 2003-06-03
US50883603P 2003-10-01 2003-10-01
PCT/US2004/017992 WO2004107963A2 (fr) 2003-06-03 2004-06-03 Systemes et procedes permettant de determiner la pression intracranienne de façon non invasive et ensembles de transducteurs acoustiques destines a etre utilises dans ces systemes

Publications (2)

Publication Number Publication Date
EP1633234A2 true EP1633234A2 (fr) 2006-03-15
EP1633234A4 EP1633234A4 (fr) 2009-05-13

Family

ID=33514059

Family Applications (2)

Application Number Title Priority Date Filing Date
EP11177783A Withdrawn EP2392262A1 (fr) 2003-06-03 2004-06-03 Procédés et systèmes pour localiser et illuminer acoustiquement une zone cible souhaitée
EP04754563A Ceased EP1633234A4 (fr) 2003-06-03 2004-06-03 Systemes et procedes permettant de determiner la pression intracranienne de fa on non invasive et ensembles de transducteurs acoustiques destines a etre utilises dans ces systemes

Family Applications Before (1)

Application Number Title Priority Date Filing Date
EP11177783A Withdrawn EP2392262A1 (fr) 2003-06-03 2004-06-03 Procédés et systèmes pour localiser et illuminer acoustiquement une zone cible souhaitée

Country Status (5)

Country Link
EP (2) EP2392262A1 (fr)
JP (1) JP2006526487A (fr)
CN (1) CN101150989B (fr)
CA (1) CA2539414A1 (fr)
WO (1) WO2004107963A2 (fr)

Families Citing this family (58)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7573627B2 (en) 2004-07-27 2009-08-11 The George Washington University Amplified bimorph scanning mirror, optical system and method of scanning
JP2006230504A (ja) * 2005-02-22 2006-09-07 Micro-Star Internatl Co Ltd 頭蓋内圧の測定方法及びシステム
US20060241438A1 (en) 2005-03-03 2006-10-26 Chung-Yuo Wu Method and related system for measuring intracranial pressure
JP4964450B2 (ja) * 2005-10-18 2012-06-27 譲二 中川原 画像処理方法、画像処理プログラム、及び画像処理装置
US8211023B2 (en) * 2006-08-11 2012-07-03 Koninklijke Philips Electronics N.V. Ultrasound system for cerebral blood flow monitoring
CN100548212C (zh) * 2007-01-18 2009-10-14 北京大学人民医院 无创颅内压监测设备
WO2008154643A1 (fr) 2007-06-12 2008-12-18 Triage Wireless, Inc. Moniteur de signaux vitaux pour mesure de tension artérielle en utilisant des formes d'onde optique, électrique et de tension
US8602997B2 (en) * 2007-06-12 2013-12-10 Sotera Wireless, Inc. Body-worn system for measuring continuous non-invasive blood pressure (cNIBP)
US11330988B2 (en) 2007-06-12 2022-05-17 Sotera Wireless, Inc. Body-worn system for measuring continuous non-invasive blood pressure (cNIBP)
US7999945B2 (en) 2007-07-18 2011-08-16 The George Washington University Optical coherence tomography / acoustic radiation force imaging probe
CN101959451B (zh) * 2008-02-27 2014-06-25 皇家飞利浦电子股份有限公司 血液动力学监测器和警报器
CA2732411C (fr) * 2008-07-30 2017-02-28 Rappaport, Arthur Procede de mesure de l'elasticite intracranienne
JP5558727B2 (ja) * 2009-02-27 2014-07-23 株式会社東芝 超音波診断装置および超音波診断装置のデータ処理プログラム
RU2540169C2 (ru) * 2009-06-09 2015-02-10 Конинклейке Филипс Электроникс Н.В. Способ и устройство для распознавания подвижных анатомических структур с использованием ультразвука
US9351707B2 (en) 2010-04-05 2016-05-31 Hitachi Aloka Medical, Ltd. Methods and apparatus to determine shear wave propagation property
JP5780748B2 (ja) * 2010-12-15 2015-09-16 株式会社東芝 医用画像処理装置
JP5346990B2 (ja) * 2011-06-03 2013-11-20 富士フイルム株式会社 超音波診断装置
US8951198B2 (en) * 2012-03-30 2015-02-10 Hitachi Aloka Medical, Ltd. Methods and apparatus for ultrasound imaging
CN103654760B (zh) * 2012-09-10 2016-08-03 焦文华 无创颅内压测量方法及应用该方法的无创颅内压分析仪
WO2014097293A1 (fr) * 2012-12-18 2014-06-26 Or-Nim Medical Ltd. Système et procédé de surveillance de la condition de débit sanguin dans une région d'intérêt dans le corps d'un patient
RU2525730C1 (ru) * 2013-03-15 2014-08-20 федеральное государственное бюджетное учреждение "Российский научно-исследовательский нейрохирургический институт имени профессора А.Л.Поленова" Министерства здравоохранения Российской Федерации Способ определения показания к хирургическому лечению нормотензивной гидроцефалии
JP6081278B2 (ja) * 2013-04-15 2017-02-15 株式会社イチカワ 頭蓋内圧測定装置及び頭蓋内圧測定方法
CN103941867B (zh) 2014-04-11 2017-07-11 北京智谷睿拓技术服务有限公司 交互方法和系统
CZ306106B6 (cs) * 2014-10-11 2016-08-03 Linet Spol. S.R.O. Zařízení a metoda pro měření intrakraniálního tlaku
EP3302234B1 (fr) * 2015-05-29 2023-09-27 Nisonic AS Dispositifs et procédé de détection de la dynamique pulsatile de la gaine du nerf optique.
US11147531B2 (en) 2015-08-12 2021-10-19 Sonetics Ultrasound, Inc. Method and system for measuring blood pressure using ultrasound by emitting push pulse to a blood vessel
CN105115592B (zh) * 2015-09-02 2018-02-16 东莞市中光通信科技有限公司 颅腔振动检测方法及装置
EP3352675A1 (fr) 2015-09-24 2018-08-01 Nihon Kohden Corporation Appareil et procédés de détection de l'augmentation du gonflement cérébral et/ou d'un changement cérébral
EP3353694A1 (fr) * 2015-09-25 2018-08-01 Institut National de la Sante et de la Recherche Medicale (INSERM) Procédé pour obtenir un modèle numérique associant une mesure objective à une sensation subjective à l'aide d'une technique d'imagerie par ultrasons et dispositif associé
EP3362178B1 (fr) * 2015-10-12 2024-01-24 Labcyte Inc. Systèmes et procédés de marquage et de caractérisation acoustique de récipients
US11589836B2 (en) * 2016-01-05 2023-02-28 Novasignal Corp. Systems and methods for detecting neurological conditions
EP3399918A4 (fr) * 2016-01-05 2019-08-21 Neural Analytics, Inc. Systèmes et procédés de détermination d'indications cliniques
KR102399223B1 (ko) 2016-06-17 2022-05-18 고쿠리츠 다이가쿠 호진 신슈 다이가쿠 두개 내압 추정방법 및 두개 내압 추정장치
EP3478184A1 (fr) * 2016-08-03 2019-05-08 PI-Harvest Holding AG Système de mesure non invasive de la pression artérielle.
JP6280997B1 (ja) * 2016-10-31 2018-02-14 株式会社Ｐｒｅｆｅｒｒｅｄ Ｎｅｔｗｏｒｋｓ 疾患の罹患判定装置、疾患の罹患判定方法、疾患の特徴抽出装置及び疾患の特徴抽出方法
JP7349727B2 (ja) * 2016-11-21 2023-09-25 センサム 生体構造の特徴づけおよび同定
JP7267917B2 (ja) * 2016-12-22 2023-05-02 サニーブルック リサーチ インスティテュート 経頭蓋超音波治療及び撮像手順を行うためのシステム及び方法
WO2018162305A1 (fr) * 2017-03-10 2018-09-13 Koninklijke Philips N.V. Dispositif de localisation et système de localisation d'un capteur acoustique
CN110392553B (zh) 2017-03-10 2023-04-04 皇家飞利浦有限公司 用于定位声学传感器的定位设备和系统
CN107296627B (zh) * 2017-07-14 2020-05-26 深圳市德力凯医疗设备股份有限公司 脑血流自动调节指数的输出方法、存储介质及超声设备
EP3740132A1 (fr) * 2018-01-19 2020-11-25 Koninklijke Philips N.V. Correction automatisée de trajet pendant une biopsie ciblée par fusion multimodale
EP3742979B1 (fr) * 2018-01-24 2022-10-19 Koninklijke Philips N.V. Imagerie ultrasonore transcrânienne guidée à l'aide de réseaux neuronaux et dispositifs, systèmes et procédés associés
JP7281210B2 (ja) * 2018-02-07 2023-05-25 サイモン メディカル アーエス 超音波血流モニタリング
US11660066B2 (en) * 2018-08-21 2023-05-30 California Institute Of Technology Wireless ultrasound monitoring device
CN109241680B (zh) * 2018-10-30 2022-03-25 南昌航空大学 基于有限元的失重状态下的人眼生物力学模拟方法
CN109730668A (zh) * 2018-11-30 2019-05-10 上海浩聚医疗科技有限公司 基于直方图估测脑自动调节能力的信号处理方法
BR112021011242A2 (pt) * 2018-12-13 2021-08-24 Liminal Sciences, Inc. Dispositivo usável, método para operar um dispositivo usável, e, aparelho
WO2020136571A1 (fr) * 2018-12-26 2020-07-02 Analytics For Life Inc. Procédés et systèmes de configuration et d'utilisation de réseaux neuronaux dans la caractérisation de systèmes physiologiques
JP7286773B2 (ja) * 2019-01-17 2023-06-05 ベラソン インコーポレイテッド ３ｄ超音波撮像を使用した定量的腹部大動脈瘤分析のためのシステム及び方法
CN113660901A (zh) 2019-04-10 2021-11-16 皇家飞利浦有限公司 颅内血压估计方法和设备
IT201900005868A1 (it) 2019-04-16 2020-10-16 St Microelectronics Srl Procedimento di elaborazione di un segnale elettrofisiologico, sistema, prodotto informatico e veicolo corrispondenti
CN110393518A (zh) * 2019-08-07 2019-11-01 西安市第四医院 一种颅内压检测系统
IT201900015926A1 (it) 2019-09-09 2021-03-09 St Microelectronics Srl Procedimento di elaborazione di segnali elettrofisiologici per calcolare una chiave virtuale di veicolo, dispositivo, veicolo e prodotto informatico corrispondenti
CN111273269B (zh) * 2020-02-18 2022-06-03 桂林电子科技大学 基于ipso-bp的频率分集阵列的雷达目标定位方法
CN111882045B (zh) * 2020-08-12 2023-10-17 北京师范大学 基于可微神经结构搜索的大脑时-空网络分解方法及系统
WO2022081907A1 (fr) * 2020-10-14 2022-04-21 Liminal Sciences, Inc. Procédés et appareil de guidage de faisceau intelligent
WO2023049529A1 (fr) * 2021-09-27 2023-03-30 Liminal Sciences, Inc. Techniques de mesure de la pression intracrânienne cérébrale, de l'élastance intracrânienne et de la pression artérielle
CN116616819B (zh) * 2023-07-25 2023-09-22 南京科进实业有限公司 一种超声经颅多普勒血流分析系统、方法及存储介质

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020082501A1 (en) * 2000-12-27 2002-06-27 Siemens Medical Systems, Inc. Method and system for control of probe heating using lens reflection pulse-echo feedback
WO2004002305A2 (fr) * 2002-07-01 2004-01-08 Allez Physionix Systemes et procedes d'evaluation non invasive d'un tissu cardiaque, et parametres s'y rapportant

Family Cites Families (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA973632A (en) 1973-05-29 1975-08-26 Arthur C. Hudson Echoencephalograph
FR2592784B1 (fr) * 1986-01-10 1992-05-07 Strauss Andreas Appareil de mesure de la pression arterielle, notamment dans l'artere ophtalmique
JPS6382623A (ja) 1986-09-27 1988-04-13 日立建機株式会社 頭蓋内圧の測定装置
US5117835A (en) 1990-07-31 1992-06-02 Mick Edwin C Method and apparatus for the measurement of intracranial pressure
JP3453415B2 (ja) 1992-12-22 2003-10-06 コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ 超音波エコーグラフィによる動脈の弾力性測定装置及び方法
US5388583A (en) 1993-09-01 1995-02-14 Uab Vittamed Method and apparatus for non-invasively deriving and indicating of dynamic characteristics of the human and animal intracranial media
US6129682A (en) 1997-02-12 2000-10-10 California Institute Of Technology Non-invasive method of measuring cerebral spinal fluid pressure
US5919144A (en) * 1997-05-06 1999-07-06 Active Signal Technologies, Inc. Apparatus and method for measurement of intracranial pressure with lower frequencies of acoustic signal
US5951477A (en) * 1997-09-11 1999-09-14 Uab Vittamed Method and apparatus for determining the pressure inside the brain
US5951476A (en) 1997-11-14 1999-09-14 Beach; Kirk Watson Method for detecting brain microhemorrhage
WO1999044504A1 (fr) * 1998-03-06 1999-09-10 Hitachi Medical Corporation Appareil video ultrasonore
US6086533A (en) 1998-06-12 2000-07-11 Children's Medical Center Corporation Non-invasive in vivo pressure measurement
US6042556A (en) 1998-09-04 2000-03-28 University Of Washington Method for determining phase advancement of transducer elements in high intensity focused ultrasound
MXPA01011471A (es) * 1999-05-10 2004-08-12 Inta Medics Ltd Monitorizacion no invasiva de la presion intracraneal.
JP3923731B2 (ja) * 1999-05-28 2007-06-06 アレツ フィジオニクス， リミテッド 血流をマッピングおよび追跡し、血流のパラメータを決定する装置および方法
EP1248556A2 (fr) * 2000-01-07 2002-10-16 Rice Creek Medical, L.L.C. Procede et dispositif non vulnerants de surveillance de la pression intracranienne
US6682491B2 (en) 2000-02-11 2004-01-27 Kci Licensing, Inc. Method for artifact reduction in intracranial pressure measurements
US6328694B1 (en) * 2000-05-26 2001-12-11 Inta-Medics, Ltd Ultrasound apparatus and method for tissue resonance analysis
WO2001097559A2 (fr) * 2000-06-15 2001-12-20 Koninklijke Philips Electronics N.V. Transducteurs ultrasonores, micro-usines, capacitifs
WO2002043564A2 (fr) 2000-11-28 2002-06-06 Allez Physionix Limited Systemes et procedes de mise oeuvre d'evaluations physiologiques non effractives
CN2514790Y (zh) * 2001-10-11 2002-10-09 李燕 无创伤颅内压监测仪
US7122007B2 (en) * 2003-05-12 2006-10-17 Caritas St. Elizabeth Medical Center Of Boston, Inc. Methods of and systems and devices for assessing intracranial pressure non-invasively
US7125383B2 (en) * 2003-12-30 2006-10-24 General Electric Company Method and apparatus for ultrasonic continuous, non-invasive blood pressure monitoring

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020082501A1 (en) * 2000-12-27 2002-06-27 Siemens Medical Systems, Inc. Method and system for control of probe heating using lens reflection pulse-echo feedback
WO2004002305A2 (fr) * 2002-07-01 2004-01-08 Allez Physionix Systemes et procedes d'evaluation non invasive d'un tissu cardiaque, et parametres s'y rapportant

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
JIANN-SHING SHIEH ET AL: "Intracranial pressure model in intensive care unit using a simple recurrent neural network through time" NEUROCOMPUTING ELSEVIER NETHERLANDS; NEW ASPECTS IN NEUROCOMPUTING: 10TH EUROPEAN SYMPOSIUM ON ARTIFICIAL NEURAL NETWORKS 2002, vol. 57, March 2004 (2004-03), pages 239-256, XP002521973 ISSN: 0925-2312 *
SCHMIDT B ET AL: "A method for a simulation of continuous intracranial pressure curves." COMPUTERS AND BIOMEDICAL RESEARCH, AN INTERNATIONAL JOURNAL AUG 1998, vol. 31, no. 4, August 1998 (1998-08), pages 231-243, XP002521971 ISSN: 0010-4809 *
See also references of WO2004107963A2 *
URSINO M ET AL: "Cerebral hemodynamics during arterial and CO(2) pressure changes: in vivo prediction by a mathematical model." AMERICAN JOURNAL OF PHYSIOLOGY. HEART AND CIRCULATORY PHYSIOLOGY NOV 2000, vol. 279, no. 5, November 2000 (2000-11), pages H2439-H2455, XP002521972 ISSN: 0363-6135 *

Also Published As

Publication number Publication date
CA2539414A1 (fr) 2004-12-16
EP1633234A4 (fr) 2009-05-13
WO2004107963A2 (fr) 2004-12-16
CN101150989A (zh) 2008-03-26
CN101150989B (zh) 2012-10-10
JP2006526487A (ja) 2006-11-24
WO2004107963A3 (fr) 2007-05-31
EP2392262A1 (fr) 2011-12-07

Similar Documents

Publication Publication Date Title
US7547283B2 (en) Methods for determining intracranial pressure non-invasively
EP2392262A1 (fr) Procédés et systèmes pour localiser et illuminer acoustiquement une zone cible souhaitée
CA2428872C (fr) Systemes et procedes de mise oeuvre d'evaluations physiologiques non effractives
US20060079773A1 (en) Systems and methods for making non-invasive physiological assessments by detecting induced acoustic emissions
US7815574B2 (en) Systems and methods for determining blood pressure
US10172527B2 (en) Method and apparatus for measuring a physical parameter in mammal soft tissues by propagating shear waves
US20060100530A1 (en) Systems and methods for non-invasive detection and monitoring of cardiac and blood parameters
US9724068B2 (en) Method for measuring intracranial elasticity
JP2005532097A (ja) 心組織およびパラメータの非侵襲的評価システムおよび方法
Imaduddin Ultrasound-based Noninvasive Monitoring Methods for Neurocritical Care

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20051227

AK Designated contracting states

Kind code of ref document: A2

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LI LU MC NL PL PT RO SE SI SK TR

AX Request for extension of the european patent

Extension state: AL HR LT LV MK

DAX Request for extension of the european patent (deleted)
PUAK Availability of information related to the publication of the international search report

Free format text: ORIGINAL CODE: 0009015

RIC1 Information provided on ipc code assigned before grant

Ipc: A61B 8/14 20060101ALI20070611BHEP

Ipc: A61B 8/12 20060101ALI20070611BHEP

Ipc: A61B 8/06 20060101ALI20070611BHEP

Ipc: A61B 8/02 20060101ALI20070611BHEP

Ipc: A61B 8/00 20060101AFI20070611BHEP

RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: SEAWALL, JASON

Owner name: THOMPSON, LEE R.

Owner name: FREDERICKSON, ROBERT C. A.

Owner name: KLIOT, MICHEL

Owner name: MOHR, BRANDT

Owner name: MOURAD, PIERRE

Owner name: UNIVERSITY OF WASHINGTON

Owner name: PHYSIOSONICS, INC.

A4 Supplementary search report drawn up and despatched

Effective date: 20090417

17Q First examination report despatched

Effective date: 20090821

REG Reference to a national code

Ref country code: DE

Ref legal event code: R003

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION HAS BEEN REFUSED

18R Application refused

Effective date: 20120504