Classifications

• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
  • A61N7/00—Ultrasound therapy
  • A61N7/02—Localised ultrasound hyperthermia
• • 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
  • A61B8/0858—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving measuring tissue layers, e.g. skin, interfaces
• • 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
  • A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
  • A61N7/00—Ultrasound therapy
  • A61N2007/0086—Beam steering
  • A61N2007/0095—Beam steering by modifying an excitation signal

Definitions

• the invention relates to medical systems and, more particularly, to non-invasive application of focused ultrasound energy to subjects such as humans, and in particular to the brain of a human subject.
• Treatment of tissues lying at specific locations within the skull may be limited to removal or ablation. While these treatments have proven effective for certain localized disorders, such as tumors, they involve delicate, time-consuming procedures that may result in destruction of otherwise healthy tissues. These treatments are generally not appropriate for disorders in which diseased tissue is integrated into healthy tissue, except in instances where destruction of the healthy tissue will not unduly effect neurologic function.
• ultrasound surgery has special appeal in the brain where it is often desirable to destroy or treat deep tissue volumes without disturbing healthy tissues.
• Focused ultrasound beams have been used for noninvasive surgery in many other parts of the body. Ultrasound penetrates well through soft tissues and, due to the short wavelengths (1.5 mm at 1 MHz), it can be focused to spots with dimensions of a few millimeters.
• heating e.g., using ultrasound, tumorous or cancerous tissue in the abdomen, for example, it is possible to ablate the diseased portions without significant damage to surrounding healthy tissue.
• the invention provides a method of delivering ultrasound signals.
• the method includes providing an image of at least a portion of a subject intended to receive ultrasound signals between sources of the ultrasound signals and a desired region of the subject for receiving focused ultrasound signals, identifying, from the image, physical characteristics of different layers of material between the sources and the desired region, and determining at least one of phase corrections and amplitude corrections for the sources depending on respective thicknesses of portions of each of the layers disposed between each source and the desired region.
• Implementations of the invention may include one or more of the following features.
• the physical characteristics are associated with material type and at least one of material density and material structure, the identifying further comprising identifying thicknesses of the layers.
• the phase corrections are determined in accordance with propagation characteristics of each of the layers. The propagation characteristics are determined based upon the material type and at least one of the material density and the material structure of each of the respective layers.
• the layers are identified using values associated with portions of the image. The values are intensities of the portions of the image.
• the phase corrections are determined using a three-layer model of a skull of the subject. Two of the three layers are assumed to have approximately identical speeds of sound, c;, therein, with the other layer having a speed of sound C;; therein, wherein the phase corrections are determined using a phase shift determined according to:
• c n is a speed of sound in the n- layer
• D n is a thickness of the n" 1 layer
• ⁇ p is a measured phase shift as a
• implementations of the invention may include one or more of the following features.
• the physical characteristics are associated with x-ray attenuation coefficients, ⁇ .
• the material between the sources and the desired region is bone.
• the phase corrections are related to the attenuation coefficient by a phase function including parameters derived at least partially experimentally. Each phase correction equals M +
• ⁇ (x) is the attenuation coefficient as a function of distance x along a line of propagation between each source and the desired region
• M, B, and C are derived at least partially experimentally.
• the amplitude corrections are related to the attenuation coefficient by an amplitude function including parameters derived at least partially experimentally. Each amplitude correction is related
• N + F ⁇ (x) + G ⁇ ( ⁇ (x)) 2 where ⁇ (x) is the attenuation coefficient as a function of distance x along a line of propagation between each source and the desired region, and where N, F, and G are derived at least partially experimentally.
• implementations of the invention may include one or more of the following features.
• the layers are identified according to both material density and material structure.
• Providing the image includes producing the image using magnetic resonance imaging.
• Providing the image includes producing the image using computer tomography.
• the sources are piezoelectric transducer elements. Both phase and amplitude corrections are determined.
• the invention provides a system for delivering ultrasound signals.
• the system includes an apparatus configured to analyze an image of at least a portion of a subject intended to receive ultrasound signals between sources of the ultrasound signals and a desired region of the subject for receiving focused ultrasound signals, the apparatus configured to determine, from the image, information about different layers of the at least a portion of the subject, and an array of sources of ultrasound signals having at least one of their relative phases and their amplitudes set in accordance with the information about each layer of the at least a portion of the subject provided by the apparatus.
• Implementations of the invention may include one or more of the following features.
• the phases are set in accordance with propagation characteristics of each layer of the at least a portion of the subject.
• the propagation characteristics are dependent upon the material type and at least one of the material density and the material structure of each layer of the at least a portion of the subject.
• the apparatus is configured to identify the layers using values associated with portions of the image. The values are intensities of the portions of the image.
• the apparatus is configured to determine the information about different layers of bone.
• the apparatus is configured to determine the phase corrections using a three-layer model of a skull of the subject. The information is
• phase corrections are related to the attenuation coefficient by a phase function including parameters derived at least partially experimentally.
• amplitude corrections are related to the attenuation coefficient by an amplitude function including parameters derived at least partially experimentally.
• implementations of the invention may include one or more of the following features.
• the system further includes a magnetic resonance imager coupled to the apparatus and configured to produce the image.
• the system further includes a computer tomography imager coupled to the apparatus and configured to produce the image.
• the sources are piezoelectric transducer elements.
• the invention provides a computer program product residing on a computer readable medium and comprising instructions for causing a computer to analyze an image of at least a portion of a subject to receive ultrasound signals between sources of the ultrasound signals and a desired region of the subject for receiving focused ultrasound signals to identify, from the image, physical characteristics of layers of material between the sources and the desired region, and to determine at least one of phase corrections and amplitude corrections for the sources depending on respective thicknesses of portions of each of the layers disposed between each source and the desired region.
• Implementations of the invention may include one or more of the following features.
• the phase corrections are determined in accordance with propagation characteristics of each of the layers.
• the propagation characteristics are dependent upon the material type and at least one of the material density and the material structure of each of the respective layers.
• the layers are identified according to both material density and material structure.
• the computer program product further includes instructions for causing a computer to produce the image using magnetic resonance imaging.
• the computer program product further includes instructions for causing a computer to produce the image using computer tomography.
• the instructions for causing a computer to identify layers of materials are for causing the computer to identify the layers of materials based upon intensities of portions of the image.
• implementations of the invention may include one or more of the following features.
• the layers are identified using values associated with portions of the image.
• the values are intensities of the portions of the image.
• the layers analyzed are layers of bone.
• the phase corrections are determined using a three-layer model of a skull of the subject. Two of the three layers are assumed to have approximately the same speed of sound, c;, therein, with the other layer having a speed of sound CJ; therein, wherein the phase corrections are determined using a phase shift determined according to: where c n is a speed of sound in the n ⁇ layer, and D n is a thickness of the n ft layer, and wherein the speeds of sound in the layers are determined according to:
• ⁇ p is a measured phase shift as a
• implementations of the invention may include one or more of the following features.
• the physical characteristics are associated with x-ray attenuation coefficients, ⁇ .
• the phase corrections are related to the attenuation coefficient by a phase function including parameters derived at least partially experimentally. Each phase correction equals M + B ⁇ (l/ ⁇ (x)) + C ⁇ (l/ ⁇ (x)) 2 , where ⁇ (x) is the attenuation coefficient as a function of distance x along a line of propagation between each source and the desired region, and where M, B, and C are derived at least partially experimentally.
• the amplitude corrections are related to the attenuation coefficient by an amplitude function including parameters derived at least partially experimentally. Each amplitude correction
• N + F ⁇ (x) + G ⁇ ( ⁇ (x)) 2 is related to N + F ⁇ (x) + G ⁇ ( ⁇ (x)) 2 , where ⁇ (x) is the attenuation coefficient as a function of distance x along a line of propagation between each source and the desired region, and where N, F, and G are derived at least partially experimentally.
• the invention provides a method of providing ultrasound signals into a subject from at least one source of an array of sources of ultrasound signals.
• the method includes (a) transmitting ultrasound energy of a selected frequency from a selected source into the subject, (b) receiving superimposed reflections of the transmitted energy, the reflections being from an outer surface of the subject and at least one interface inside the subject, (c) repeating (a) and (b) using ultrasound energy of frequencies other than the selected frequency, (d) determining a frequency difference between frequencies associated with relative extrema of the received reflections, and (e) using the determined frequency difference and a thickness, of at least a portion of material between the selected source and a desired region in the subject for receiving focused ultrasound energy signals, to determine a phase correction for the selected source.
• Implementations of the invention may include one or more of the following features.
• the method further includes (f) providing an image of at least a portion of a subject intended to receive ultrasound energy signals between sources of the energy signals and the desired region, and (g) identifying, from the image, the thickness of at least a portion of material between the selected source and the desired region.
• the method further includes repeating (a) - (e) for each of the sources other than the selected source.
• the phase correction is determined according to:
• ⁇ is the phase correction
• f is a frequency to be transmitted
• d is the thickness
• the invention provides logic for use in a system for providing ultrasound energy into a living subject from an array of sources of ultrasound energy signals.
• the logic is configured to control apparatus to (a) transmit ultrasound energy of a selected frequency from a selected source into the subject, (b) receive superimposed reflections of the transmitted energy, the reflections being from an outer surface of the subject and at least one interface inside the subject, (c) repeat (a) and (b) using ultrasound energy of frequencies other than the selected frequency, (d) determine a frequency difference between frequencies associated with relative extrema of the received reflections, and (e) use the determined frequency difference and a thickness, of at least a portion of material between the selected source and a desired region in the subject for receiving focused ultrasound energy signals, to determine a phase correction for the selected source.
• Implementations of the invention may include one or more of the following features.
• the logic is further configured to cause the apparatus to (f) provide an image of at least a portion of a subject intended to receive ultrasound energy signals between sources of the energy signals and the desired region, and (g) identify, from the image, the thickness of at least a portion of material between the selected source and the desired region.
• the logic is further configured to cause the apparatus to repeat (a) - (e) for each of the sources other than the selected source.
• the logic is configured to cause the apparatus to determine the phase correction according to:
• ⁇ is the phase correction
• f is a. frequency to be transmitted
• d is the thickness
• the invention provides a method of delivering ultrasound signals.
• the method includes providing an image of at least a portion of a subject intended to receive ultrasound signals between sources of the ultrasound signals and a desired region of the subject for receiving focused ultrasound signals, identifying, from the image, physical characteristics of portions of material between the sources and the desired region, and using a spectral propagation analysis to determine at least one of phase corrections and amplitude corrections for the sources depending on the respective physical characteristics of the portions of the material disposed between each source and the desired region.
• Implementations of the invention may include one or more of the following features.
• the physical characteristics include density and thickness.
• the identifying identifies the physical characteristics of different layers of the portions of the material.
• the physical characteristics include angle of a layer interface relative to a reference.
• Implementations of the invention may also include one or more of the following features.
• the spectral propagation analysis includes a wavevector-frequency domain projection algorithm.
• the using includes comparing projected phases from the sources to the desired region accounting for the material and ignoring the material.
• the using assumes perpendicular incidence of ultrasound.
• the method further includes determining an incident angle of ultrasound to the material. The identifying identifies the physical characteristics of different layers of the portions of the material and the determimng an incident angle determines incident angles for each layer of the material.
• the invention provides a system for delivering ultrasound signals.
• the system includes an apparatus configured to analyze an image of at least a portion of a subject intended to receive ultrasound signals between sources of the ultrasound signals and a desired region of the subject for receiving focused ultrasound signals, the apparatus configured to determine, from the image using a spectral projection analysis, information about the at least a portion of the subject, and an array of sources of ultrasound signals having at least respective ones of their relative phases and their amplitudes set in accordance with the information about the at least a portion of the subject provided by the apparatus.
• Implementations of the invention may include one or more of the following features.
• the apparatus comprises means for identifying physical characteristics of multiple layers of the portion of the subject, the physical characteristics including density and thickness.
• the physical characteristics further include incident angles of the layers relative to respective sources, the apparatus further comprising means for comparing projected phases of the signals from the sources to the desired region accounting for the portion of the subject and ignoring the portion of the subject.
• the apparatus comprises means for performing a wavevector-frequency domain projection algorithm.
• the invention provides a computer program product residing on a computer readable medium and including computer-readable, computer executable instructions for causing a computer to analyze an image of at least a portion of a subject to receive ultrasound signals between sources of the ultrasound signals and a desired region of the subject for receiving focused ultrasound signals to identify, from the image, physical characteristics of respective portions of material between the sources and the desired region, and to determine, using a wavevector-frequency domain projection algorithm, at least one of phase corrections and amplitude corrections for the sources depending on respective physical characteristics of portions of the material disposed between each source and the desired region.
• Implementations of the invention may include one or more of the following features.
• the physical characteristics include density and thickness.
• the instructions for causing the computer to analyze cause the computer to identify the physical characteristics of different layers of the portions of the material.
• the physical characteristics include angles of layer interfaces relative to the sources.
• Implementations of the invention may also include one or more of the following features.
• the instructions for causing the computer to analyze cause the computer to compare projected phases from the sources to the desired region accounting for the portions of the material and ignoring the portions of the material.
• the computer program product further includes instructions for causing the computer to determine the projected phases accounting for the portions of the material assuming perpendicular incidence of ultrasound.
• the computer program product further includes instructions for causing the computer to determine the projected phases accounting for the portions of the material by determining and using incident angles of ultrasound relative to the layers of the portions of the material.
• the instructions for causing the computer to determine incident angles cause the computer to determine incident angles for each layer of the portions of the material.
• Ultrasound can be focused accurately within an intact skull, e.g., for ultrasound therapy.
• Different skulls e.g., different skull thicknesses, densities, and/or structures, can be accommodated for ultrasound therapy.
• Real-time adjustments to ultrasound therapy can be made. Effects on phase and/or amplitude of energy passing through bone (or other tissue) may be determined and used to compensate the phase and/or amplitude of energy applied to the bone (or other tissue).
• Incident angles of ultrasound relative to a skull can be accounted for and excitation characteristics of ultrasound transducers adjusted accordingly.
• a wavevector-frequency domain projection algorithm can be used to determine ultrasound transducer excitation characteristics for providing ultrasound non-invasively into a skull. Correction factors for ultrasound array transducers can be determined in an efficient, timely manner.
• FIG. 1 is a schematic diagram of an ultrasound therapy system according to the invention.
• FIG. 2 is a 3-dimensional rendering of a portion of a patient's skull.
• FIG. 3 is a flow diagram of a process of determining excitation correction factors and exciting transducer elements using the determined factors.
• FIG. 4 is a schematic diagram of another ultrasound therapy system according to the invention.
• FIG. 5 is a flow diagram of a process of obtaining phase shift factors using the system shown in FIG. 4.
• FIG. 6 is schematic diagram showing planar field rotation in a wavevector- frequency domain, defining a new projection plane.
• FIG. 7 is a schematic ray diagram of ultrasound propagation through non-parallel layers of different-density materials.
• Energy such as ultrasound energy
• a subject such as a human or animal
• Arrays of radiating transducer elements may be used to transmit energy into the subject, and the amplitudes of the signals transmitted by the elements can affect how much energy penetrates the subject and the relative phases and amplitudes of energies transmitted can help to focus a distribution of energy in the subject.
• the phases and amplitudes of ultrasound signals may be affected/distorted by many causes such as different material properties of materials through which the signals are propagated. For example, different types of materials within a patient, e.g., bone, muscle, and fat, have different propagation and attenuation constants of ultrasound energy.
• phase and/or attenuation of ultrasound signals may affect the phase and/or attenuation of ultrasound signals. Because the signals from different transducer elements may encounter different thicknesses and contours of materials and possibly air-filled or liquid- filled pockets between transducer elements and a region to be imaged/treated, the phases of the signals from the transducer elements will often be distorted. The resulting phases and/or amplitudes will therefore often be different than desired if the transmission phases and/or amplitudes are not compensated for this distortion.
• Embodiments of the invention provide techniques for compensating for phase distortions and attenuation variances when treating a patient using ultrasound.
• imaging techniques such as computer tomography (CT), magnetic resonance imaging (MRI), etc.
• properties such as thickness, density, and structure of materials are determined. It has been discovered that each of these properties affects the phase distortion of materials, so that phase corrections determined using all three properties will be better than those determined using only one or two of the properties.
• the determined properties for materials in a subject between transducer elements and a region to be treated are inserted into formulas developed by the inventors.
• the formulas use the determined properties and known characteristics, e.g., propagation speed, to calculate phase adjustments for each of the transducer elements.
• x-ray attenuation can be related to phase distortion and acoustic attenuation. Formulas relating to x-ray attenuation to phase distortion and acoustic attenuation can be used to compensate for such variances by adjusting transmitted phase and amplitude accordingly.
• an ultrasound therapy system 10 includes an imager 12, a phased array 14 of n transducer elements 16, a signal adjuster 18, a controller 20, and a frequency generator 22.
• the system 10 is configured to determine characteristics of a skull 28 of a patient 30, and to apply ultrasound energy (e.g., in the range 0.01 MHz to 10 MHz) that is focused in the patient's brain.
• Signals to the array are provided by a driving arrangement similar to that reported in Daum et al., "Design and Evaluation of a Feedback Based Phased Array System for Ultrasound Surgery.” IEEE Trans. Ultrason. Ferroelectr. Freq. Control 45(2):431-4, 1998, but with a driving frequency selected between about 0.1 MHz and about 10 MHz.
• the power and phase to each transducer element 16 may be manually controlled or automatically controlled using software and feedback.
• the array 14 of transducer elements 16 is configured to be disposed on or near the external surface of the patient's skull 28.
• the array 14 is configured in a curved shape (e.g., spherical, although sections of other shapes are possible such as planar) conducive for being placed on the patient's head and for focusing ultrasound energy at a distance from the surface of the array 14.
• the transducer elements 16 of the array 14 are piezoelectric transducer elements arranged in the array 14 as shown.
• the transducer elements e.g., 1 cm 2 piezoelectric ceramic pieces
• the array 14 may be formed from one or more pieces of piezocomposite material, or any material that converts electrical energy to acoustic energy.
• the transducer elements 16 may be configured for electrical resonance at 50 ⁇ to help match input connector impedance.
• the array 14 is coupled to the signal adjuster 18 that is further coupled to the frequency generator 22.
• the frequency generator 22 is configured to provide a common radio frequency (RF) signal as the input signal to the signal adjuster 18.
• the radio frequency generator 22 can be of any type that will produce the appropriate signals for the signal adjuster 18.
• the generator 22 may be a Model DS345 generator available from Stanford Research Systems.
• the radio frequency generator 22 and signal adjuster 18 are configured to drive the individual transducer elements 16 of the array 14 at the same frequency, but at different phases (and possibly different amplitudes), in order to transmit ultrasound energy through the patient's skull 28 and focus the energy at a selected region within the patient's brain.
• the generator 22 is coupled to the adjuster 18 to split the generator's output signal to provide n input signals to the signal adjuster 18.
• phase shifters 26 Coupled to receive each of the n input signals from the frequency generator 22 are n pairs of amplifiers 24r24 n and associated phase shifters 26 ⁇ -26 n of the signal adjuster 18. Each pair of phase shifter 26 and amplifier 24 represents a channel of the signal adjuster 18.
• the phase shifters 26 are configured to provide n independent output signals to the amplifiers 24 by altering or adjusting the phase (and possibly amplitude) of the
• phase shifters 26 provide approximately 1 degree precision (8-bit resolution, although lower phase resolution may be adequate for many applications).
• the amplifiers 24 24 n are configured to amplify the signals from the phase shifters 26 and to provide the amplified signals to the transducer elements 16 through connections, e.g., coaxial cables, individually connecting the amplifiers 24 and the transducer elements 16.
• An internal power meter is configured to monitor power supplied by the amplifiers 24.
• phase shift factors ⁇ i-oc,- of the phase shifters 26 provide steering of the ultrasound beam absent an object in the path of the ultrasound energy and also provide compensation for phase distortion in the ultrasound output by each transducer element 16 induced by the patient's skull.
• the component of each phase shift factor associated with steering can be computed using known techniques. The component of each phase shift
• phase shift factor ⁇ i-ot n associated with phase distortion compensates for perturbations and distortions introduced by the skull 28, the skin/skull interface, the dura matter/skull interface, by variations in the skull thickness and by structural considerations such as air- filled or liquid-filled pockets in the skull 28.
• the two components that make up the phase shift factors ⁇ i-o are summed in order to determine the composite phase shift factors ⁇ i-
• ⁇ n for the respective channels in order to focus ultrasound energy at a desired steering angle relative to, and distance from, the array 14.
• the phase shift factors ⁇ i-ot n are provided by the controller 20.
• the controller 20 is logic that may be provided by software, hardware, firmware, hardwiring, or combinations of any of these.
• the controller 20 can be a general purpose, or special purpose, digital data processor programmed with software in a conventional
• the controller 20 is configured to determine
• the phase shift factors ⁇ , ⁇ - ⁇ n as described below based on information obtained from the imager 12 as indicated by arrow 32.
• the information includes the thickness and density of the patient's skull 28 for each portion of the skull 28 between each transducer element 16 and the desired focal point in the patient's brain.
• Information from the imager 12 is conveyed directly from the imager 12 because the imager 12 is configured to automatically analyze images and determine characteristics of interest from the images. Enough information is provided by the imager 12 to the controller 20 to
• the controller 20 is configured to manipulate images from the imager 12.
• the controller 20 is configured to produce a 3-dimensional rendering of the patient's skull 28 from 2-dimensional images received from the imager 12 and to determine skull thickness from the 3-dimensional rendering.
• the 3-dimensional rendering can be divided by the controller 20 into voxels (a volume pixel of the 3-dimensional image).
• the imager 12 is configured to obtain images of the interior of the patient's head, and in particular images that provide information regarding thickness, density, and structure of bone of the patient's skull 28.
• the imager 12 may be a Magnetic Resonance Imaging (MRI) device or Computer Tomography (CT) device.
• the imager 12 is configured to scan the patient's skull 28 and provide information related to skull thickness, density and structure. This information includes 2-dimensional images of varying intensity from which 3-dimensional renderings can be made and from which thicknesses and densities can be determined and/or inferred. Three-dimensional image acquisition may also be possible and can be used.
• the imager 12 can determine and provide the CT number (also called Hounsfield number) for each pixel in images provided by the imager 12.
• the skull 28 includes two layers 50, 54 of trabecular bone and a layer 52 of cortical bone.
• a scan direction of the imager 12 is defined as the Cartesian x-axis so that the image plane is defined by the y-axis and z-axis. From the 3-D rendering, two vectors 34, 36 on the surface 38 of the skull 28 are determined using the 3 rd nearest neighboring points on the surface in the x and y directions from the point of interest rn on the skull's surface 38.
• the controller 20 is configured to use the vectors 34, 36 to calculate a vector 40 that is normal to the surface 38.
• the controller can calculate a scalar product of the vector 40 and a vector 42 that is the direction of propagation of energy from a transducer element 16 x . This scalar product is the incident angle:
• a Cartesian (normalized) unit vector n (u,v,w) may be calculated by the controller 20
• the three layers 50, 52, 54 can be treated separately to determine phase compensation due to the thicknesses D l5 D 2 , D 3 of the layers 50, 52, 54 as part of a three- layer model.
• the skull 28 consists of individual homogeneous layers.
• the speed of sound is assumed to be 2.5 x 10 3 m/s for the central layer 52 and 2.9 x 10 3 m/s for the inner and outer layers 50, 54.
• the expected phase shift across the skull 28 using this three-layer model is:
• each 0.15 mm 2 voxel is assigned an intensity value. It is assumed that the intensity is linearly proportional to bone density and density is scaled to MKS units using air and water in the image as reference intensities.
• Mean intensity is determined by summing the CT intensity values along the axis of propagation 42 inside the bone and dividing by the total number summed of voxels.
• the voxels may include air-filled or liquid-filled pockets. The sound speed for such voxels is assumed to be the speed of sound in water, or if air-filled, then complete reflection of the ultrasound can be assumed.
• Error due to skull density has been calculated as the difference between the measured phase and that given by Eq. (3) (for a single-layer model).
• An empirical correction factor has been obtained by fitting (using a polynomial curve fit) percent error as a function of the mean intensity. The correction factor is:
• the controller 20 is configured to fit the speed as a function of density according to:
• Eq. (5) is used by the controller 20 to find the speed of sound values.
• two sound speeds are calculated. These two speeds are the speed Cj for the trabecular layers and the speed C ⁇ of the cortical (central) bone. Given the thicknesses
• the speed cn of the cortical layer is fit by the controller 20 as a function of density according to:
• the controller 20 uses information from the imager 12 to adjust the phase of the transducer elements 16.
• the imager 12 takes images of the patient's skull 28. This may be done remotely in space, e.g., in another hospital, from the controller 20.
• the imager transmits information, e.g., intensities of portions of the image, related to skull thickness, density, and structure to the controller 20. This transmission can be separated in time from when the imager 12 takes the image and can also be performed by human intervention, e.g., by recording the images on a CD and replaying the CD in the controller 20.
• the controller 20 manipulates this information to determine thicknesses of layers, and to identify the layers, of the patient's skull 28 by analyzing intensities of portions of the image.
• the controller 20 manipulates this information to determine thicknesses of layers, and to identify the layers, of the patient's skull 28 by analyzing intensities of portions of the image.
• the controller 20 manipulates this information to determine thicknesses of layers, and to identify the layers, of the patient
• controller 20 determines excitation correction factors such as the phase shift factors ⁇ i-ct n
• the frequency generator supplies energy to the adjuster 18.
• the adjuster 18 adjusts the phase of the energy from the frequency generator 22
• the adjusted energies are sent to the transducer elements 16 that convert the energies into ultrasound waves and transmit these waves into the patient's skull 28.
• Images from the imager 12 can be supplied in real time and the amplitude and/or phasing of the energy applied to the patient 30 changed in response to the images.
• the power supplied to the patient's skull 28 depends on the type of therapy. For ablation, approximately 2-3 kW for approximately 10 seconds using 64 transducer elements may be used. If more transducer elements 16 are used, then less total power may be used and vice versa. For opening the blood-brain barrier, about 100 times less power than for ablation may be used due to preformed gas bubbles in the area of interest. The ablation power can also be reduced by the preformed gas bubbles. Using bursts of energy has been found to reduce, if not eliminate, affects on phase due to standing waves that may occur if the transducer elements 16 constantly emit energy.
• the array 14 of transducer elements 16 shown in FIG. 1 may contain fewer transducer elements 16 than shown.
• the phase shift factors ⁇ i-c n may be pre-stored in the channels of signal adjuster 18 instead of being provided by the controller 20.
• functions described as being performed by the controller 20 could be performed by the imager 12 and vice versa, or by a person using the system 10, e.g., calculating densities and providing input data to the controller 20 regarding phase shift factors ⁇ i-o ⁇ .
• a controller 92 is similar to the controller 20 (FIG. 1), but is also configured to control switches 94 ⁇ -94 n such that the transducer elements 16 are connected to the controller 92 in a receive mode and to the adjuster 18 in a transmit mode.
• the controller 92 controls the switches 94 to be positioned such that one switch, e.g., switch 94j, is in the transmit mode.
• the frequency generator 22 produces a low-frequency ultrasound wave train of about 5-30 cycles. Some of this wave train reflects off of the skull 28 and some of the wave train passes through the skull 28 into the patient's brain.
• the controller 92 causes the switch 94 1 to switch into the receive mode, connecting the transducer element ⁇ 6 ⁇ to the controller 92.
• the wave train reflects off of the skull 28 and reflections are received by the transducer element 16 ⁇ and recorded by the controller 92.
• the propagation speeds and the pulse train lengths are such that energy from a pulse train is still being transmitted into, and reflected by, the outer surface of the skull 28 when energy from that pulse train that was reflected by the inner surface of the skull 28 passes through the skull 28 toward the array 14.
• the received reflections of the inner and outer surfaces are superimposed.
• stage 76 a query is made as to whether a series of frequencies to be transmitted has been completed. If not, then the process 70 returns to stage 72 to transmit a different frequency in the series. If the series has been completed, with corresponding recorded amplitude data as a function of frequency, then the process 70 proceeds to stage 78.
• the controller 92 analyzes and processes the recorded information to
• the controller 92 deconvolves the recorded frequency response that includes the superimposed reflections from the interfaces at the inner and outer surfaces of the skull. The superimposed reflections produce a periodic appearance of local maxima and minima in the data as a function of frequency. Information from images, e.g., CT or MRI images, are used by the controller 92 to determine the locations of the inner and outer surfaces, and thus the skull thickness. The controller 92 estimates the phase shift ( ⁇ ) using the distance, ⁇ f, between like extrema,
• ⁇ 2 ⁇ f [ (d/c 0 ) - (l/(2 ⁇ f)) ] (9) where f is a frequency to be transmitted for which the phase correction is to be used, d is the skull thickness determined from the image, and c 0 is the speed of sound in water. Adjustments may be made to this phase shift using density correction techniques. The process 70 is repeated for the remainder of the transducer elements 16.
• Phase shifts and acoustic attenuation due to transskull sonication can also be determined from analysis of CT images. For purposes of determining phase shift and acoustic attenuation, it can be assumed that there is essentially a one-to-one correlation
• Equations relating x-ray attenuation to phase and acoustic attenuation have been developed as discussed below, with the equations including parameters determined from fitting to experimental data.
• Experimental data has been determined by applying ultrasound to a pig skull sample using a 1" transducer element made by Panametrics of Waltham, Massachusetts, operating at 730 kHz and positioned about 7 cm from a 0.6mm hydrophone. The skull was attached to a 3-D positioning system and the transducer element was driven by a 10-cycle sinusoid signal produced by a waveform generator such as a Wavetek model 395 made by Wavetek Ltd. of Norwich, United Kingdom, or a waveform generator made by, e.g., Fluke Corporation of Everett, Washington.
• a waveform generator such as a Wavetek model 395 made by Wavetek Ltd. of Norwich, United Kingdom, or a waveform generator made by, e.g., Fluke Corporation of Everett, Washington.
• the hydrophone response was recorded on an HP54602 digital oscilloscope made by Hewlett Packard of Palo Alto, CA. Measurements of the skull sample were obtained by immersing the sample in water and using dual-slice spiral acquisition with 0.5 mm slices. Phase shift through the skull can be related to x-ray attenuation by assuming a formulaic relationship and fitting actual data to the formula. The x-ray attenuation
• A 2 ⁇ f(A' - l/c 0 )
• B 2 ⁇ fB'
• C 2 ⁇ fC
• p 2 , pi are the first and last pixels within the bone along the line of propagation and the sums are performed within these boundaries.
• the values of p 2 , pi are obtained manually from intensity profiles drawn along the lines of propagation. These values may also be obtained automatically using well-known edge-detection techniques.
• the equation for ⁇ can be modified to include a thickness-independent
• Eq. 13 can be modified to:
• ⁇ A(p2-p ⁇ +l) + B ⁇ (l/ ⁇ (x)) + C ⁇ (l/ ⁇ (x)) 2 + J (14)
• J is the phase shift at zero thickness and its non-zero value is probably due to experimental error.
• Acoustic attenuation through the skull can also be related to x-ray attenuation by assuming a formulaic relationship and fitting actual data to the formula.
• the acoustic attenuation through bone is given by:
• I and I 0 are the acoustic intensities before and after the passage through the bone
• ⁇ AC (x) is the local acoustic attenuation coefficient.
• ⁇ AC (x) is the local acoustic attenuation coefficient.
• the imager 12 supplies the CT numbers of images to the controller 20 that uses the CT numbers and the experimentally-determined values for A, B, C, J, E,
• the system otherwise operates generally as described above, with the adjuster 18 adjusting phase and amplitude of energy from the generator 22 as indicated by the controller 20.
• the system 10 shown in FIG. 1 can be configured to apply a projection algorithm to identify phase and/or amplitude adjustments to be applied by the amplifiers 24 and/or the phase shifers 26.
• the algorithm is based on a layered wavevector- frequency domain model that propagates ultrasound from a hemispherical transducer array through the skull 28 and uses input from CT scans of the skull 28.
• the projection algorithm, that the controller 20 is configured to perform calculates driving phases and/or amplitudes for each transducer 16 in the array 14 to improve, an possibly maximize, the signal at an intended focus. Experimental results have shown that this technique can be used for completely non-invasive ultrasound brain therapy, including surgery.
• the projection algorithm as implemented by the system 10, and especially the controller 20, concentrates on a noninvasive focusing method that uses CT images to predict how the ultrasound will propagate through the skull 28.
• the algorithm predicts the behavior of the ultrasound field after passing through the skull bone 28, which causes significant reflection, diffraction and absorption of the field.
• the system 10 accurately obtains knowledge of the thickness and internal structure of the skull bone 28 and achieves precise registration between the skull 28 and the ultrasound array 14.
• the algorithm provides a relatively uncomplicated, computationally-feasible mechanism for determining phase and/or amplitude corrections without oversimplification.
• the projection algorithm uses thickness, density, and orientation information obtained from CT images of the head 28. This information is entered into a propagation model that can propagate ultrasound through arbitrarily oriented (non parallel) layers. Operating in the wavevector-frequency domain, this model rapidly propagates sections of the ultrasound field through the skull 28. Two specialized cases of the model are examined for the present focusing algorithm. The first case propagates ultrasound through the skull 28 with normal incidence, and assuming that the skull is homogeneous. The second case considers the angle between the skull surface and the incident acoustic field. Both of these models have been tested using ex vivo human skulls and hydrophone measurements in a water bath.
• the Projection Algorithm The Projection Algorithm
• a projection algorithm in the wavevector-frequency domain is used to determine propagation through the skull 28 and correction factors for the ultrasound array transducers 16.
• the projection algorithm assumes that the acoustic field satisfies the linearized Stake's equation,
• the field recorded in a plane ZQ is thus related to the field at any other plane z by a simple transfer function given in the right hand side of Eq. (22).
• the pressure field in space at z may be obtained by taking the inverse Fourier transform of the equation.
• Propagation in wavevector space can be performed using Eq. (22) after the spatial Fourier transform of the field is calculated for a given plane.
• This planar cross section of the field, z 0 is chosen to be planar and perpendicular to the propagation axis.
• the field can be projected to an arbitrary new plane z.
• the array 14 is sectioned into small areas that are individually propagated through the skull 28 in a manner that, as described below, yields more easily in Cartesian space.
• the controller 20 divides the array 14 into segments equal to the array's transducer elements (e.g., 280 mm 2 ), although this does not have to be the case. For example, a large element could be segmented into several smaller elements.
• the controller 20 calculates the acoustic pressure in water due to each segment over a plane located directly above the skull's outer surface. Using the backward projection approach, the controller 20 first calculates the field of each transducer element 16 at the transducer's geometric focus in water. The field at the focus is calculated using the Rayleigh-Somerfeld integral,
• the controller 20 preferably calculates this field only once for a particular transducer, and uses it repeatedly. Alternatively, the effects of imperfections on the array could be added by replacing Eq. (23) with physical measurements of the field over this area for each element 16 in the array 14.
• the field due to the n-th element is Fourier transformed by the controller 20 giving a planar field p ⁇ k x , k y , ⁇ , z 0 ) .
• This planar field is rotated in the wavevector-
• each point describes a planar wave with frequency ⁇ passing through the measurement plane at an angle equal to
• the ⁇ ⁇ amplitude of the wave is equal to the modulus of p .
• FIG. 6 illustrates this concept for a
• each temporal frequency possesses a unique angular spectrum, forming a three-dimensional space consisting of two spatial frequency dimensions and one temporal frequency dimension.
• the tested signals are harmonic and may be expressed in two dimensions.
• each point p in wavevector space is mapped to the rotated plane such that:
• the controller 20 obtains the angles directly from the Fourier transform of the pressure field, taking angles
• Propagation through layered media in wavevector space can be achieved by determining exactly how the layers distort each point in the space. For an arbitrary field incident upon a series of planar layers, each wavevector component over the plane at z 0 is
• the transfer function may readily be written in a closed form.
• the unit vectors normal to the layer surfaces In addition to the thickness across the z-axis, z n , the sound speed c n , and density of each layer p n , the unit vectors normal to the layer surfaces
• h n are calculated. For a given initial wavevector k 0xy , the ray path from (0,0, zo) between any two surface interfaces traverses a distance of
• r is the vector extending along the layer from z-axis to the
• the transmitted wavevector on the right hand side of Eq. (28) may be obtained by crossing both sides of the equation with n n+l . Using cross product relations, it may be
• the incident unit wavevector of the n* layer is equal to the transmitted wave of the (n-l) s
• the spatial phase at the plane z is related to the ray phase at N by
• ⁇ N (k Nxy ,o)) ⁇ R ⁇ k Nxy , ⁇ ) - 2 ⁇ k N r N sin ⁇ Nxy , as illustrated in FIG. 7.
• the pressure over the plane at z can be expressed in terms of the ray phase presented in Eq. (31) and the transmission coefficient in Eq. (26) by
• the layered-projection method may be used for the propagation of a field through a curved surface, provided the surface is sufficiently smooth relative to the highest relevant wavenumber.
• the field is projected to a plane near the surface and is divided into a series of virtual sources.
• the region of the surface penetrated by a given source is approximated as planar, with the actual surface over the beamwidth of the source preferably agreeing with this approximation to at least within %. of the maximum frequency & max .
• the source diameter S is preferably large enough that
• the signal and z is the distance from the starting plane to the surface.
• the planar field is uniquely divided into a continuous series of sources, the fields of the sources will overlap on the surface.
• each of the surfaces is segmented as described above. It is possible that components with high transmission angles will leave the surface of interest.
• multiple layers of the skull 28 can be taken into account by determining the transmission angle through each layer, using this angle and layer thickness and orientation
• Standard techniques may be used to register the array 14 and images of the skull 28. Registering the array 14 and the images helps to ensure that the proper focal point for the ultrasound is achieved. Proper registration helps ensure that the proper portion of the skull 28 is being analyzed to yield the accurate thicknesses, densities, orientations (e.g., incident angles), and locations of layers of the skull 28 between the array 14 and the desired focal point within the skull 28. With accurate measurements, the controller 20 can determine accurate phase and amplitude corrections for the amplifiers 24 and the phase shifters 26 to focus ultrasound from the array 14 at a desired location inside the skull 28.
• Standard techniques for registering the array 14 and the skull images include using a reference frame that attaches to the array 14 and to the skull 28 and that has reference markers that can be identified in CT and MR images.
• the reference markers can be used to rotate CT and MR images to register the images (with CT images yielding bone characteristics, while MR images may be used during therapy due to better soft- tissue resolution).
• CT images yielding bone characteristics while MR images may be used during therapy due to better soft- tissue resolution
• one or both of the CT and MR images may be rotated until they match.
• Other techniques that yield registration so that the array's orientation relative to the skull 28 can be determined are also acceptable.
• Eq. (33) Density measurements obtained by analyzing the CT images are input to Eq. (33) to provide the effective skull sound speed values in the focusing algorithm.
• Eq. (33) is exemplary only, and not limiting.
• Eq. (33) is based on experimental data, and thus dependent on the acquired data. Further, other relationships of sound speed and density are possible, and acceptable, including other linear data fits. Also, other fits, e.g., non-linear fits, of sound speed as a function of skull density may be used.
• the acoustic pressure is calculated by the controller 20 at the intended focal point for each element 16 after propagation through the skull 28.
• the controller 20 can perform its calculations under at least two different scenarios. In the first case (Case 1), the controller 20 ignores incident angles and assumes the ultrasound is normally incident upon the skull 28. In the second case (Case 2), the controller 20 uses the incident angle ⁇ i determined from Eq. (26) over the skull surface to include incidence in the model. The pressure phase is compared with the phase expected if the skull 28 were not present. The change in phase caused by the skull 28 is recorded and used for correcting the driving phase of the transducer array 14. The driving phase of each element 16 is adjusted by an amount
• the imager 12 takes a CT image of the skull 28. From this image, the controller 20 determines locations, orientations, thicknesses, and densities of layer portions between respective array elements 16 and a desired focal point inside the skull 28. The controller 20 can assume that the skull 28 is a single layer or can take into account different skull layers. From the determined skull data, the controller 20 determines lengths traveled by ultrasound from the elements 16 through the skull 28 toward the focal point. The controller 20 applies relationships of phase and amplitude to density, either taking into account incident angles or assuming planar propagation. The controller 20 thus determines propagation speed and attenuation (e.g., using Eq.
• the controller 20 determines phase and amplitude correction factors to be applied by the phase shifters 26 and the amplifiers 24 to yield focused ultrasound at a desired location inside the skull 28.
• the entire projection algorithm was implemented in Matlab, on a series of Pentium and AMD-based computers over an NT network.
• a typical processing time for producing a focus within the skull 28 for 320 elements 16 using a 1 GHz machine with 256 MB of RAM was approximately five hours.
• the transducer array was a 30-cm-diameter, 0.74 MHz hemisphere divided into 500 individual elements. Only the lower 320 elements of the array were used for the measurements in order to allow a larger range in the movement of the skull.
• a reference frame was attached to the skull to assist with registration of the array and the skull.
• a hydrophone was placed inside the skull to measure the ultrasound inside the skull from the array.
• the peak signal without phase correction was found to occur at a mean distance of 1.1 mm from the intended focal location with values ranging from 0.45 mm to 2.5 mm and a standard deviation (STD) of 0.64 mm.
• STD standard deviation
• the main focus was scattered into two or more foci, so the peak signal refers to that of the largest (larger) focus.
• the phasing algorithm was used, a single focus was always produced, regardless of whether the incident angle was considered.
• the mean distance from the peak was reduced to 0.59 mm with values ranging from 0.22 mm to 1.1 mm and a STD of 0.26 mm. With the angle of incidence included in the algorithm, the mean distance was further reduced to 0.48 mm over a range of distances from 0.20 mm to 0.82 mm and a STD of 0.22 mm.
• Peak voltage-response-squared measurements were also observed to improve with the phasing algorithm. Without phasing, the average peak was found to be 34 % of the value obtained by the invasive hydrophone-phasing technique with a variation from 14 % to 58 % over the samples. The lower amplitudes corresponded to skulls that scattered the field into multiple foci.
• the phasing algorithm without correction for incident angle, improved the peak value on average to 46% of the hydrophone-phased signal with a range between 22% and 58%. When the incident angle was included, the average value was 45% with a range from 29% to 59%.
• the peak values between the two models were, on average, similar. However, the variance was reduced when incident angles were included.

Landscapes

• Health & Medical Sciences (AREA)
• Life Sciences & Earth Sciences (AREA)
• Biomedical Technology (AREA)
• Animal Behavior & Ethology (AREA)
• Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
• Veterinary Medicine (AREA)
• Radiology & Medical Imaging (AREA)
• Engineering & Computer Science (AREA)
• Public Health (AREA)
• General Health & Medical Sciences (AREA)
• Heart & Thoracic Surgery (AREA)
• Molecular Biology (AREA)
• Surgery (AREA)
• Medical Informatics (AREA)
• Biophysics (AREA)
• Physics & Mathematics (AREA)
• Pathology (AREA)
• Ultra Sonic Daignosis Equipment (AREA)

Applications Claiming Priority (7)

Publications (2)

Family

ID=27405745

Family Applications (1)

Country Status (2)

Families Citing this family (14)

Citations (1)

Family Cites Families (2)

• 2002
  • 2002-08-26 WO PCT/US2002/027130 patent/WO2003017843A1/en active Application Filing
  • 2002-08-26 EP EP02796433A patent/EP1429659A4/de not_active Withdrawn

Patent Citations (1)

Non-Patent Citations (4)

Also Published As

Similar Documents

Legal Events