WO1994004073A1 - Blood flow monitoring system - Google Patents

Blood flow monitoring system Download PDF

Info

Publication number
WO1994004073A1
WO1994004073A1 PCT/US1993/008105 US9308105W WO9404073A1 WO 1994004073 A1 WO1994004073 A1 WO 1994004073A1 US 9308105 W US9308105 W US 9308105W WO 9404073 A1 WO9404073 A1 WO 9404073A1
Authority
WO
Grant status
Application
Patent type
Prior art keywords
means
substrate
system
device
sensor
Prior art date
Application number
PCT/US1993/008105
Other languages
French (fr)
Inventor
Glenn W. Laub
Original Assignee
Zertl Medical Inc.
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

Links

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/42Details of probe positioning or probe attachment to the patient
    • A61B8/4272Details of probe positioning or probe attachment to the patient involving the acoustic interface between the transducer and the tissue
    • A61B8/4281Details of probe positioning or probe attachment to the patient involving the acoustic interface between the transducer and the tissue characterised by sound-transmitting media or devices for coupling the transducer to the tissue
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/06Measuring blood flow
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01FMEASURING VOLUME, VOLUME FLOW, MASS FLOW OR LIQUID LEVEL; METERING BY VOLUME
    • G01F1/00Measuring the volume flow or mass flow of fluid or fluent solid material wherein the fluid passes through the meter in a continuous flow
    • G01F1/66Measuring the volume flow or mass flow of fluid or fluent solid material wherein the fluid passes through the meter in a continuous flow by measuring frequency, phaseshift, or propagation time of electromagnetic or other waves, e.g. ultrasonic flowmeters
    • G01F1/663Measuring the volume flow or mass flow of fluid or fluent solid material wherein the fluid passes through the meter in a continuous flow by measuring frequency, phaseshift, or propagation time of electromagnetic or other waves, e.g. ultrasonic flowmeters by measuring Doppler frequency shift

Abstract

A blood flow monitoring system including a monitor (2), at least one sensor (48) securable to a patient's skin, and an electrical cable (46) connecting the sensor to the monitor. The monitor generates an electrical signal corresponding to a first acoustic frequency, processes electrical input signals corresponding to a second (doppler-shifted) acoustic frequency, and displays a patient's blood flow characteristics based upon the input signals. The monitor may determine and display a patient's blood flow characteristics based on an idealized normalized absolute (INAF) scale or a pulse palpitation equivalents (PPE) scale. The sensor preferably includes a substrate (66A) having circuit elements (68A) thereon for conducting electricity, at least one transducer element (72) for transmitting the output signals to a patient's blood vessel and receiving input signals reflected therefrom. The sensor is releasably secured by adhesive (100) or the like, to the patient's skin and is preferably disposable, i.e. detachably connected to the cable (46).

Description

TITLE

BLOOD FLOW MONITORING SYSTEM

FIELD OF THE INVENTION

The present invention relates to medical detection and monitoring equipment and, more particularly to non-invasive apparatus for detecting and monitoring conditions indicative of a patient's blood flow characteristics.

BACKGROUND OF THE INVENTION

Monitoring of the peripheral pulses is an integral part of the clinical evaluation of patients after many surgical and diagnostic procedures. Variations in these patients' peripheral pulses may signify important changes in blood flow which, if not promptly detected, can have serious consequences.

During and after certain types of therapeutic and diagnostic procedures, there is an increased risk that blood flow through an artery may be compromised. Presently, blood flow monitoring typically involves periodic clinical assessment, possibly supplemented by use of an audio Doppler ultrasonic unit. Clinical blood flow assessment alone can be highly subjective due to the individualized nature of each clinician's interpretations when palpating a patient's pulse. Even when Doppler monitoring is utilized, it is problematic because, due to the periodicity of clinical assessments, a clinician must try to position the blood flow sensor in the same location where the last measurement was taken, angle the probe identically, and make a subjective evaluation of the blood velocity represented by the sound and decide if it has changed significantly since the last measurement.

An advantage exists, therefore, for a blood flow monitoring system which provides consistent blood flow monitoring in a reproducible manner whereby changes in profusion may be promptly and accurately identified so as to enable a clinician to be timely and reliably informed of a patient's clinical status.

SUMMARY OF THE INVENTION

The present invention provides a system including a transcutaneous, non-invasive, two channel Doppler blood velocity detector or monitor which utilizes the Doppler ultrasonic principle to detect blood flow in a blood vessel such as an artery and continuously displays detected blood flow as a numerical value on a digital readout. The monitor is capable of simultaneously detecting blood flow in two blood vessels. The invention provides a substantial advantage over known systems in that a significant change in the displayed blood flow value timely alerts the clinician to immediately perform standard clinical and hemodynamic evaluations.

Although one Doppler ultrasonic sensor may be sufficient, the patient being monitored by the subject system is typically connected to the monitor using a pair of securable Doppler ultrasonic sensors. One sensor is positioned over the appropriate artery in each selected limb and securely attached with an adhesive strip. The sensor is connected into the main monitor unit which contains the electronics to process and display the blood flow in each artery or other blood vessel. After the patient is connected to the monitor, blood flow is continuously detected and displayed. Any change in the monitored arterial blood flow alerts the nurse or doctor to immediately reevaluate the patient's vascular status using standard clinical and hemodynamic evaluations.

In addition, the system according to the present invention employs novel sensor constructions and electronic signal processing techniques which operate in concert to decrease the subjective and uncertain nature of evaluations obtained from traditional Doppler stethoscopes. The monitor thus allows for continuous detection and quantification of arterial blood flow velocities. Two channels are available for blood flow evaluation and the results can be presented both visually and audibly. Two digital displays allow quantified average blood flow to be displayed while associated vertically oriented light bars provide a qualitative indication of instantaneous flow. Functions are selected directly from front panel switches making the monitor simple to operate.

The digital readout displays blood flow in two scales. In one scale, the blood flow is displayed in units of flow. Blood flow computations are based on

Doppler measurements of blood velocity and assumed physiological parameters based on idealized assumptions coded to the selected sensor. These flow measurements are not absolute blood flow rates, but correlate to and parallel changes in the blood flow through the measured artery. In the other scale, the blood flow is displayed as an integer from 0 to 4. This scale corresponds to the standard clinical pulse scale used by clinicians as determined by palpation.

The monitor also can present the Doppler audio signal in two forms. Either the standard Doppler audio signal or a synthesized tone can be selected. In the synthesized tone mode the frequency of the tone is proportional to blood flow. The monitor has an internal speaker for audio output and a headphone jack is preferably provided should the clinician desire to privately listen to the patient's blood flow pattern through optional headphones.

Monitoring begins as soon as the monitor is operational and the sensors are positioned. Functional modes are selected directly from the front panel. Error conditions are indicated on the digital displays.

The monitor is normally powered from a hospital grade, 120V/60 Hz AC power supply. A detachable hospital grade three-wire power cord with a three-contact plug is preferably provided. The instrument can also operate, however, from an internal rechargeable battery power pack. The battery backup supply continuously charges whenever the AC power cord is plugged in, regardless of the position of the POWER switch. The battery backup supply automatically activates whenever primary 120V AC power is disconnected and the power switch is in the ON position. Under fully charged conditions, the battery will power the instrument in excess of one hour.

The internal battery pack provides secondary power to monitor, during power failure and patient transportation. The battery indicator on the front panel turns on when the battery supply is active and flashes when the supply is low. The condition of the battery can be checked by removing the AC power cord and turning the instrument on.

The presently preferred monitor contains two channels of processing and display capabilities. A preferred measurement method is to use a continuous wave

(CW) acoustic Doppler transducer, although pulsed or phrased array Doppler transducers may also be employed. The Doppler shifted return from the insonified blood vessel is processed to determine the instantaneous blood flow velocity therein. The instantaneous flow rate corresponding to this velocity is computed, based on the lumen diameter. The lumen diameter is encoded into the transducer and, for purposes of error analysis, the diameter is assumed to be accurate. At present, 16 choices of lumen diameter appear to be sufficient to achieve the objects of the present invention. It will be appreciated, however, that the monitor may accommodate more or less than 16 lumen diameter selection possibilities. The computation also assumes a parabolic velocity profile across the lumen of the vessel. This measurement does not determine the direction of blood flow, but only its magnitude. It is assumed that blood flow will be undirectional in all cases of interest.

The instantaneous flow rates obtained by the above method are summed up (integrated) during the course of a single heartbeat to obtain an aggregate flow rate. At completion of these computations, the display is updated with new information, using either the Idealized Normalized Absolute Flow (INAF) or Pulse Palpitation Equivalents (PPE) scale. INAF is computed according to the encoded transducer parameter relating to lumen diameter (e.g. , choice 1 of 16) so that the display will represent a typical flow rate for the artery being monitored. PPE is derived from the INAF figure based on a simple table lookup. Beat to beat averaging is performed to stabilize the display.

Preferably, as described above, a continuous wave acoustic probe insonifies the vessel and the Doppler shifted return signal is processed to determine blood flow. It is primarily the red blood cells that scatter the Doppler frequency ultrasonic from the blood. The ultrasonic is reflected from each scatterer (red blood cell) with a change in frequency given by the Doppler equation:

Fd = 2FQ • (v • cos φ /c)

where: F, = Doppler frequency

F = probe frequency v = velocity of the scatterer φ = ultrasonic beam to vessel angle c = speed of sound

Thus by measuring the Doppler shifted frequency velocity of the scatterer can be determined. Flow rate can then be computed by multiplying the velocity by the cross-sectional area of the vessel:

Fr = v • A

In the continuous wave Doppler situation, continuous ultrasonic is radiated toward the vessel from the transmitting transducer. The ultrasonic is backscattered from the blood with a shift in frequency and picked up by a receiving transducer. The total returned signal is the sum of the contributions from the different scatterers within the acoustic beam. Because the different blood elements have different velocities, a spectrum of Doppler shifted frequencies is received. This spectrum characterizes total blood flow, assuming a fairly even concentration and distribution of red blood cells.

The receiving transducer signal is amplified and coherently demodulated against the carrier frequency to bring the Doppler shift frequencies to base-band. The resulting spectrum is in the audio band (0 - lOKHz) and is processed by additional circuitry to provide a waveform whose amplitude tracks the peak frequency of the spectrum. This is directly proportional to the peak velocity as described by the Doppler equation. The 'velocity' waveform is sampled by an internal computer and instantaneous flow is computed based on the cross-sectional area of the vessel. The flow is integrated (summed) over a cardiac cycle to produce normalized average flow.

The basic flow model used for computation assumes a straight, round vessel of known diameter. The velocity profile is assumed to be parabolic, and the blood flow unidirectional. The beam to vessel angle is assumed to be fixed in the sensor unit. It is understood that the above assumptions are ideal and only approximate actual clinical conditions.

The present invention thus provides consistent blood flow monitoring in a reproducible manner whereby changes in profusion may be promptly and accurately identified so as to enable a clinician to be timely and reliably informed of a patient's clinical status.

Other details, objects and advantages of the present invention will become apparent as the following description of the presently preferred embodiments and presently preferred methods of practicing the invention proceeds.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will become more readily apparent from the following description of preferred embodiments thereof shown, by way of example only, in the accompanying drawings, wherein: Figure 1 is a perspective view of a preferred embodiment of a blood flow monitor of the blood flow monitoring system of the present invention;

Figure 2 is a view of a front panel of a preferred embodiment of a blood flow monitor adapted for use in the blood flow monitoring system of the present invention;

Figure 3 is a view of a rear panel of a preferred embodiment of a blood flow monitor adapted for use in the blood flow monitoring system of the present invention;

Figure 4 is a view similar to Figure 1 depicting the monitor in a first operational mode;

Figure 5 is a view similar to Figure 4 illustrating the monitor in a further operational mode;

Figure 6 is a view of a sensor cable adapted for connection to the blood flow monitor of the blood flow monitoring system of the present invention;

Figure 7 is a view representing a setup arrangement of the monitoring system of the present invention;

Figure 8 is a plan view of one side of a printed circuit board which, forms a component a preferred embodiment of a blood flow sensor adapted for use in the monitoring system of the present invention; Figure 9 is a view showing a first preferred embodiment of a blood flow sensor adapted for use in the blood flow monitoring system of the present invention including a first preferred arrangement of transducer elements mounted to the printed circuit board of Figure 8;

Figure 10 is an enlarged elevation view of a preferred embodiment of a transducer element adapted for use in a sensor of the blood flow monitoring system of the present invention;

Figure 11 is an enlarged plan view of a portion of Figure 9 revealing the electrical connection of a transducer element to the printed circuit board;

Figure 12 is an elevation view of the arrangement shown in Figure 11;

Figure 13 is an end elevation view of a further preferred embodiment of a blood flow sensor adapted for use in the blood flow monitoring system of the present invention;

Figure 14 is an end elevation view of a further preferred embodiment of a blood flow sensor adapted for use in the blood flow monitoring system of the present invention;

Figure 15 is a section view of a further preferred embodiment of a blood flow sensor adapted for use in the blood flow monitoring system of the present invention;

Figure 16 is a plan view of the printed circuit board of the sensor of Figure 15; Figure 17 is a section view of a further preferred embodiment of a blood flow sensor adapted for use in the blood flow monitoring system of the present invention;

Figure 18 is a plan view of the printed circuit board of the sensor of Figure 17;

Figure 19 is a section view of a further preferred embodiment of a blood flow sensor adapted for use in the blood flow monitoring system of the present invention;

Figure 20 is an end elevation view of the blood flow sensor of Figure 19 shown in contact with the surface of a patient's skin as when monitoring a patient's blood flow characteristics;

Figure 21 is an end elevation view of a further preferred embodiment of a blood flow sensor adapted for use in the blood flow monitoring system of the present invention;

Figure 22 is a section view of a further preferred embodiment of a blood flow sensor adapted for use in the blood flow monitoring system of the present invention;

Figure 23 is a plan view of a further preferred embodiment of a blood flow sensor adapted for use in the blood flow monitoring system of the present invention;

Figures 24 through 41 depict a plurality of interrelated flow diagrams which graphically represent the presently preferred process by which the monitor of the blood flow monitoring system of the present invention determines the blood flow characteristics of a patient being monitored; and

Figure 42 depicts a schematic diagram of a portion of the circuitry arrangement of the monitor of the blood flow monitoring system of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

Referring to Figure 1, there is shown a presently preferred embodiment blood flow monitor, herein designated by reference numeral 2, which forms an essential component of the blood flow monitoring system of the present invention. The monitor 2 includes a housing having a plurality of walls 4 a front panel 6 and a rear panel 8 (Figure 3).

Figure 2 shows that, among a number of controls, displays and other user-interactive components, the front panel includes a two-position power switch 10 which is selectively positionable into an ON and an OFF position for activating and deactivating the monitor 2. Following activation of the monitor, there is a brief initialization period wherein all display lights and indicators (to be later described) briefly turn on and monitoring begins. The monitor also contains internal batteries (not illustrated) which charge whether the power switch 10 is in the ON or the OFF position. Moreover, unplugging of the monitor will not cut the supply of power thereto. That is to say, the batteries automatically power the instrument if the power switch is left in the ON position while the unit is unplugged. A battery indicator light 12 illuminates when the monitor 2 is operating under battery power, i.e., any time the instrument is unplugged, the power switch 10 is ON, and the batteries are sufficiently charged. The battery indicator light 12 flashes when the energy level in the batteries passes below a threshold level of charge, such as, for example, when the batteries have about five minutes of effective power remaining.

A first, vertically oriented, illuminated bar meter or bar graph display 14 is provided on the left side of the front panel 6 and will hereinafter be referred to as the "right" bar graph. A second similar bar graph display 16 is provided on the right side of the front panel and will hereinafter be referred to as the "left" bar graph. Right bar graph 14 and left bar graph 16 assume their ostensibly inappropriate names and locations on panel 6 for the following reason. During blood flow monitoring operations, the front panel of monitor 2 is typically arranged so as to face in a direction generally, toward the clinician and away from the patient. As such, the flow characteristics of an artery or other blood vessel being monitored on a right side of a patient's body will be displayed on the left side of the front panel 6 as, for example, at right bar graph 14, and vice versa with regard to a blood vessel being monitored on a left side of the patient's body. The advantage of the arrangement is that the front panel displays, from the clinician's visual perspective, are in alignment with the orientation of the patient's body. That is to say, from the clinician's vantage point, the "right" and "left" displays (including bar graphs 14 and 16 as well as digital displays 18 and 20 to be described hereinbelow) are situated in a direct rather than inverse positional relationship with respect to a patient's body. Thus, the clinician is not hampered by the potentially confusing situation wherein the displays, if located on the front panel according to standard convention, would be in mirror image or inverse positional relationship vis-a-vis the patient's body. The right and left bar graph displays 14 and 16 provide continuous indication of information borne by the Doppler signal. The type of information displayed by the right and left bar graphs, however, is determined by the operational mode of the monitor as selected by the clinician through manipulation of switches 28, 30 and 32 which operational modes and switches therefor are described later herein. In BOTH mode, the right bar graph 14 displays the instantaneous peak Doppler frequency detected in the received Doppler signal for the right channel which the left bar graph 16 provides a similar display for the left channel. In RIGHT mode, the right bar graph 14 displays the signal strength at the Doppler signal to aid the clinician in placement of a "right" blood vessel flow sensor, whereas in LEFT mode, the left bar graph 16 displays similar information to assist the clinician in placement of a "left" blood vessel flow sensor.

The front panel 6 is also provided with a "right" digital display 18 and a "left" digital display 20 which are respectively located generally on the left and right sides of the front panel for the reason discussed previously in connection with right and left bar graphs 14 and 16. The right and left digital displays 18 and 20 display quantitative information corresponding to detected blood flow for either or both the right and left channels.

In BOTH mode, the right digital display 18 displays blood flow information for the right channel and the left digital display 20 ^displays the same information for the left channel. In RIGHT or LEFT mode, the right digital display 18 displays blood flow for the selected channel

(right or left) and the left digital display 20 provides the pulse rate for the selected right or left channel.

According to a presently preferred embodiment, the monitor 2 is operative so as to update the digital displays 18 and 20 once every heartbeat or pulse or once a second if no detectable pulse is present. In FLOW mode the digital displays possess a display range of 1 to 1999 that is representative of estimated blood flow in units of ml/min. In PULSE mode, a plus sign (+) precedes the data which has an integer range of 0 to 4 and which is a qualitative scale based on the standard clinical peripheral pulse scale used by clinicians, determined by palpation.

The digital displays 18 and 20 provide numeric information about blood flow. They also give the user mode and operating condition information. During normal operation the digital displays are updated continuously. Each channel is processed independently and continuously so that changing modes only affects what is displayed. If no pulse is detected the displays 18 and 20 will automatically update at one second intervals. If the pulse rate exceeds 140 beats/minute, display updates occur every second pulse.

Reference numerals 22 and 24 designate right and left channel connectors or sensor jacks to which an appropriate patient cable (to be later described in connection with Figures 6 and 7) is connected so as to enable blood flow monitoring or selected patient blood vessel(s) .

An output jack 26 is preferably provided for receipt of the plug of an optional headphone as would be required, for example, when background noise is excessive and the clinician wishes to aurally monitor the patient's blood flow pattern.

Three push button switches 28, 30 and 32 enable the clinician to select the desired monitoring mode. Each mode is selected by depressing the appropriate mode switch. There are three possible operating modes: RIGHT (deployable by pressing switch 28), BOTH (deployable by depressing switch 30) and LEFT (deployable by depressing switch 32). A light contained within each button indicates which mode is active.

When BOTH mode is selected, the monitor will display both the left and right channels flow information with no audio output. This mode allows for dual artery blood flow monitoring. LEFT and RIGHT modes are used for single channel monitoring and only the selected channel is displayed. Audio output for that channel is enabled. Only one mode can be active at a time.

A rotatable volume knob 34 adjusts the volume or both a built-in speaker 40 (Figure 3) and the headphone output. Audio output is possible only in RIGHT or LEFT channel mode.

A depressible FLOW/PULSE switch 36 permits the clinician to select display of the detected blood flow velocity in flow units (FLOW) or as a clinical correlation with the standard peripheral pulse scale (PULSE). In PULSE mode a plus sign (+) precedes an integer value of 0 through 4. The FLOW/PULSE switch 36 has a push-on push-off actuation and a change from FLOW to PULSE mode is signified by the plus sign on the appropriate digital display.

As mentioned hereinabove, two different types of audio output are available. They are selected by a push-on push-off AUDIO SELECT switch 38. When the switch 38 is in the out position, raw Doppler audio frequencies are present. When the switch 38 is depressed, a computer processed audio signal is produced. The processed signal is a tone which occurs once per pulse period and the tone duration is proportional to the pulse period. The tone pitch is proportional to the blood velocities. And, as mentioned above, audio output is only active in single channel mode, i.e. , either RIGHT OR LEFT.

Turning to Figure 3, there is shown the speaker

40 situated in rear panel 8 from which audio output typically originates, unless a suitable headphone plug is inserted into headphone jack 26 which serves to override audio output from the speaker 40. An AC receptacle 42 allows the monitor 2 to be connected via a suitable hospital grade power cord (not illustrated) to a suitable source of 120 V AC power. The main fuse for the monitor is located in a snap out drawer 44 which is associated with the AC power receptacle 42.

Turning to Figure 4, there is shown a typical monitor display when the monitor is operating in the BOTH mode.

In BOTH mode, the Left 20 and Right 18 digital displays show blood flow for the left and right channels, respectively. As mentioned previously, if the monitor is additionally in FLOW mode, the digital displays have a range of 1 to 1999 representing estimated blood flow in the units of ml/min. ; and if in PULSE mode, a plus sign (+) precedes the data which has a integer range of 0 to 4, a qualitative scale based on the standard clinical peripheral pulse scale used by clinicians, determined by palpation.

The numbers that appears on the digital displays in FLOW mode is the instantaneous flow averaged over time. The display is updated with each pulse. In BOTH mode a six pulse or greater average is desirably used to remove display jitter from minor pulse-to-pulse deviations. Thus, it may take the display several pulse beats to stabilize following changes in the patient's status or the sensor's position. If the displays are unstable, the sensor location is checked to ensure good signal strength and reliable operation. Poor signal quality can deleteriously affect the flow computation resulting in flow rate fluctuations.

The bar graph displays 14 and 16 provide a visual indication of the instantaneous peak Doppler blood velocity detected for each channel. This is a useful indicator of proper sensor positioning and instrument operation. A strong pulse should result in at least 1/2 of the bar graphs to be illuminated. A partially lit but unchanging display indicates flow with no pulsatility, while a unit display signifies lack of velocity information for flow computation. Audio output is disabled in BOTH mode.

Figure 5 represents a typical monitor display when the monitor is operating in the LEFT mode. However, the following description refers to both LEFT and RIGHT modes since the functions of these modes are identical except for the fact that they are associated respectively with either left or right channel monitoring operations.

LEFT and RIGHT channel modes are used primarily for sensor positioning for the left and right channel sensors, respectively. During LEFT or RIGHT mode, the right digital display 18 shows flow information and the left digital display 20 shows the pulse rate in beats/minute for the selected channel. The flow data displayed in LEFT or RIGHT mode is preferably averaged over two pulses to allow rapid response during sensor positioning. However, in either of these modes, the output of the digital displays 18 and 20 will fluctuate somewhat more than when the monitor is in the BOTH mode. The respective bar graph 14 or 16 of the selected channel displays the received Doppler power and is used for optimizing sensor placement while other channel's bar graph is deactivated. The audio output is enabled during RIGHT and LEFT modes and can be selected as raw Doppler audio or computer enhanced audio using the AUDIO SELECT switch 38.

Figure 6 generally schematically depicts a sensor cable 46 which forms a component of the blood flow monitoring system of the present invention and which carries at a first end thereof a sensor 48 (various preferred embodiments of which will be described in greater detail hereinafter) and at its other end with a connector 50.

Figure 7 reveals the manner in which the blood flow monitoring system of the present invention is set up so as to monitor the blood flow of a patient's dorsal pedal artery.

The clinician first connects a connector 52 at one end of a patient cable 54 to one of the right or left channel sensor jacks 22 or 24 (in the present example, the right channel sensor jack) provided in the front display panel 6 by pushing the connector 52 in the direction of arrow 56. The other end of the patient cable is then secured, as by adhesive tape 58, or the like, at a location near to the monitoring site. Thereafter, the connector 50 of the sensor cable 46 is inserted in the direction of arrow 60 into mating connection with an appropriate connector 62 carried by the patient cable.

To optimize sensor placement and, hence, detected Doppler signal strength, it is suggested that the clinician first palpate the dorsal pedal pulse of the desired limb (in this case the right foot) and then mark the strongest palpable pulse site with a non-allergenic felt tip pen, or the like.

The foregoing procedure is then repeated if it is desired to monitor blood flow in another limb such as, for example, the left foot.

The instrument is then turned on by depression of power switch 10 into the ON position and the volume knob 34 is set to approximately mid-range. The appropriate LEFT or RIGHT mode switch 28 or 32 is depressed, the FLOW/PULSE switch 36 is set to FLOW mode and the AUDIO SELECT switch 38 is set to AUDIO.

The sensor cable 46 is then secured to the patient with adhesive tape 64 approximately six inches from the monitoring site. The sensor 48 is then aligned wit the pen marks earlier provided on the skin and the viability of the site is assessed by listening for a crisp, pulsating audio signal. If the monitoring room is particularly noisy or if the pulse is weak, headphones may be used for improved listening. The sensor 48 is moved slowly back and forth over the marked area while the clinician listens to the audio signal in conjunction with observing the peak reading or the appropriate bar graph 14 or 16 until the detected signal strength is maximized.

Once the optimal pulse site is located, the sensor 48 is held firmly in place while carefully removing adhesive strip covers attached thereto whereupon the sensor is pressed into adhesive contact with the target pulse site. These pulse site locating steps are of course repeated for the second sensor if two blood vessels are to be simultaneously monitored. The monitor 2 then begins to automatically display, depending upon the selected mode of operation, the flow characteristics of the blood vessel(s) being monitored.

According to the present invention, the sensor 48 may assume an assortment of preferred embodiments, a number of which are illustrated in Figures 8-23. For purposes of distinguishing the various embodiments of the sensor 48, each sensor separately described hereinafter will be designated by, in addition to reference numeral 48, an alphabetic character, e.g. , 48A, 48B, etc. Although any of the sensors disclosed herein may be suitably employed for general blood vessel monitoring, each embodiment has benefits of application that vary somewhat from the other, e.g. , a specific embodiment may be particularly well adapted for monitoring a specific class and/or diameter of blood vessel. Further, each sensor desirably includes an encoding element to identify its type whereby the monitor interrogates the sensor during power-up or when the sensor is attached and calibrates its computations to the type of sensor being used.

Figures 8 and 9 demonstrated a first preferred embodiment of blood flow sensor, designated by reference numeral 48A, for use in the blood flow monitoring system of the present invention. It will be understood that in all embodiments, the sensors are preferably disposable and are constructed for releasable attachment to the sensor cable 46 (Figures 6 and 7).

Sensor 48A comprises a circuit board 66A which, like all of the circuit boards to be described herein, may be formed of any suitable material including, but not limited to, ceramic material, substantially rigid molded plastic, flexible plastic strips and paper-like products. The circuit board 66A is provided with a plurality of electrically conductive traces or circuit elements 68A and contacts 70A which may be applied to the board 66A via printing with conductive epoxies, metal sputtering processes or other suitable techniques. Sensor 48 may also comprise an integrated circuit design, e.g., VLSI chips with integrated transducer crystals.

In accordance with this particular embodiment, the sensor 48A carries two transducer elements or crystals 72, one of which operates as a Doppler signal transmitter element and the other as a Doppler signal receiver element whereby the sensor operates as a continuous-wave (CW) transducer in which the Doppler signal is continuously transmitted and received and thus continuously processed and monitored by monitor 2.

The preferred general construction of a transducer crystal 72, which is adaptable for use in all sensor embodiments disclosed herein, is revealed in Figure 10. The crystal is desirably formed of a piezoelectric material 74 with layers of conductive material 76 provided on each surface thereof and an electrical contact 78 provided on at least one of the layers of conductive material. Also, it may be formed as part of an integrated circuit.

Each crystal 72 is preferably adhesively secured to the board 66A by means of conductive epoxy or the like. Similar material, as indicated by reference numeral 80, is used to secure a conductive element such as a wire 82 to the electrical contact 78 provided on the exposed or upper surface of the crystal 72 and its corresponding contact 70A provided on board 66A as in the manner shown in Figures 11 and 12. According to this preferred embodiment, each crystal 72 is placed so as to lie substantially coplanarly with the other crystal. Moreover, sensor 48A, like all embodiments of sensors pursuant to the present invention, is preferably provided with other electrical circuitry components such as, for example, resistor 84, to enable encoding of the sensor. As may be appreciated in the preferred embodiment utilizing an integrated circuit design, it is possible to eliminate the separate adhesive step.

In some Doppler frequency ranges, sensor performance may be enhanced by angling one or both of the crystals 72 of a two crystal transducer system such that the crystals do not lie in coplanar relationship. Figures 13 through 20 reflect a number of presently preferred strategies for achieving an acceptable non-coplanar arrangement of the crystals.

According to Figure 13, wherein the sensor is designated by reference numeral 48B and comprises a suitable circuit board 66B, one of the crystals 72 is shown to be secured (as by conductive epoxy or the like) generally parallel t the plane or the surface of the circuit board much as the crystals 72 are secured to the previously described sensor 48A. The other crystal, however, is illustrated as being disposed at an inclination with respect to both the first crystal and the surface of the circuit board 66B. To achieve this effect, a small shim 86 my be positioned beneath one edge of the crystal. The shim 86 may assume the form of a wire, a pinched metal spacer, a "thick" circuit trace, or any other suitable means for achieving the desired degree of inclination.

Figure 14 is similar in many regards to the sensor embodiment 48B shown in Figure 13 except that in this embodiment the sensor, reference 48C, includes suitable shim means 86 provided under appropriate edges of both crystals 72 such that the crystals are in non-coplanar and substantially reverse image relationship with one another and are not parallel to the surface plane of the circuit board 66C to which they are secured.

The embodiment of the sensor 48D shown in Figure 15 achieves essentially the same non-coplanar relationship between the crystals 72 as that represented by the sensor 48B of Figure 13. However, instead of providing shim means under an edge of one of the crystals 72, an inclined notch or cavity 88 is provided in the circuit board 66C which receives one of the crystals. Again, both crystals are preferably secured by conductive epoxy to the circuit board 66D which epoxy extends through plated through-bores 90 to traces or circuits (not illustrated) that are preferably carried on the opposite side of the circuit board. A plan view of the notched circuit board 66D is shown in Figure 16.

The sensor 48E shown in Figure 17 is similar in principal to sensor 48D of Figure 15 except that two notches 88 are machined, formed or otherwise satisfactorily provided in one surface of the circuit board, reference 66E, to achieve the desired inclination in each crystal 72. As with board 66D, board 66E is preferably provided with plated through-bores 90 for enabling electrical communication of the conductive epoxy which secures the crystals to the board with unillustrated tracers at the opposite side of the board. A plan view o the doubly notched circuit board 66E is shown in Figure 18.

Figure 19 depicts a sensor 48F according to a further preferred embodiment of the present invention. Like board 66E of Figures 17 and 18, board 66F of Figure 19 includes two notches 88 for receiving crystals 72 therein. Unlike board 66E, however, the notches 88 are cut so as to incline generally in the same direction rather than in substantial reverse image to one another. Further, the notch or notches 77 in board 66F (and, boards 66D and 66E) may be inclined at any suitable angle with respect to the blood vessel being monitored. A presently preferred angular range for such notches is between about 30° to about 60° and the notches, if two are present, may incline at the same or at different angles relative to one another. Note, for example, that the arrows 92 and 94 extending from the crystals 72 in Figure 19 emphasize the difference in relative angular inclination of the notches 88 provided in board 66F.

Figure 20 demonstrates the sensor 48F in contact with the surface of a patient's skin 96 such as when monitoring the blood flow characteristics in a blood vessel such as artery 98. Note that the sensor 48F, like all of the sensors described herein (although not earlier illustrated), is provided with adhesive means 100 for securing the sensor to the patient's skin. Such adhesive means 100 may suitably comprise, for example, an adhesive foam pad, or the like.

In addition, although circuit boards 66D, 66E and 66F are illustrated as having through-bores 90 to facilitate electrical communication between the crystals 72 at one side of the boards and the unillustrated circuit traces at the opposite sides thereof, it will be understood that the traces may be provided on the same side of the boards as the crystals (such as is depicted in Figures 8-14), thereby obviating the necessity for the through-bores. The primary advantage of the through-bores 90, however, is that manufacture of the notched circuit boards 66D, 66E and 66F is simplified since placement of traces across the notched areas is eliminated. It is also contemplated, and in some circumstances desirable, to use a pulsed rather than a continuous wave Doppler sensor in the blood flow monitoring system of the present invention. Examples of such pulsed sensors, which represent further preferred sensor embodiments, are shown in Figures 21 and 22. The primary difference between these sensors and those previously described herein is that a single crystal serves as a transducer. That is, the crystal both transmits the Doppler signal to and receives the reflected Doppler signal from the blood vessel being monitored.

Turning first to Figure 21, the sensor, designated by reference numeral 48G, includes a circuit board 66G which includes traces and contacts similar to those shown in Figure 8 except that only such traces and contacts as are required for a single crystal are provided. Again, the crystal 72 is secured to the traces and electrically joined to appropriate contacts by way of conductive epoxy or similar material.

Figure 22 is similar to Figure 21 except that a notch 88 and through-bores 90 are provided in the circuit board 66H of the sensor 48H in the manner discussed hereinabove in connection with the descriptions of boards 66D, 66E and 66F.

Figure 23 represents a further preferred embodiment of sensor, herein designated by reference numeral 481, which t is adapted for beneficial use in the blood flow monitoring system of the instant invention. Sensor 481 is a phase array Doppler transducer system in which several, preferably between about three to about one thousand or more crystals 72 are adhesively secured to the one side of the circuit board 661. The board may be a multi-layer board with driver and/or decoder elements incorporated therein. Element 100 represents a suitable connector for joining the board to the sensor cable 46 (Figures 6 and 7). A sensor of this particular construction finds unique advantage when computer imaging of the blood flow characteristics of the monitored blood vessel is desired.

In all of the above-described sensors, at least some of the circuitry is an integral part of the substrate which supports the transducer elements or crystals 72. In conventional sensor designs, the substrates, i.e., the boards, to which the crystals are mounted are independent of the sensor circuitry. In other words, in known sensors the crystals themselves carry the necessary circuitry whereas the boards are merely substrates for supporting the crystals.

Although not exhaustively illustrated to exhibit all of its possible manifestations, it will be appreciated that the above-described embodiments of the sensor 48 are merely indicative of a number of the presently contemplated embodiments thereof and that specific disclosure of these embodiments should not be construed to be limitative of the sensor designs which may reasonably lie within the scope of the present invention.

The following is a discussion of the operation of the blood flow monitor 2, described primarily by way of an algorithmic representative of a presently preferred software system incorporated therein, by means of which the monitor performs its essential functions.

The software system is an integral pat of the instrument and provides an intelligent interface between the processing hardware and the user, typically a doctor, nurse or other qualified clinician. The software system processes and interprets the Doppler signal information received by the sensor(s) 48 and enables the monitor 2 to display the information in a user selectable format via the front panel 6 displays. The software system further performs system status monitoring and error handling functions.

Preferably, the software is written in the C programing language and is desirably run on an Intel 80C196 micro-controller. The code is compiled into 80C196 machine language and stored in EPROM memory for execution. Once compiled, there can be no variations to the system performance or operation without source code modification and recompilation (other than by the front panel 6 controls). Execution begins when the power switch 10 is turned ON and continues until the power switch is turned OFF or the batteries become too depleted for operation.

The software system is comprised of various modules or tasks. These tasks operate at either the foreground or background processing level. The foreground processing tasks perform the critical real time operations that are synchronous to an internal clock. They are invoked by interrupts that happen at precise intervals and are executed immediately regardless of what is happening in the background. The foreground tasks consist of a set of interrupt handlers that perform data sampling and control the feedback loop of the velocity estimator filter. Data processing in the foreground is preferably held to a minimum.

The background tasks process the data in a less time critical fashion, dealing with each new piece of data as it arrives and processing it to completion. Most of the processing work occurs at the time a blood flow pulse is detected, thereby allowing pulse processing to be spread out over several sample periods. The two levels of processing are isolated from each other by a "first in first out" data buffer, or FIFO. The foreground task writes the data samples into the FIFO when they are sampled and the background task takes them out in order and processes them. The FIFO has sufficient capacity to hold all the samples of a pulse, allowing a latency of up to one pulse period for new display information.

The background system is set up so that all tasks stem from one main procedure. On powerup, the main control module/loop executes the system initialization and goes into a loop waiting for data in the input FIFO. When data arrives the main loop dispatches it to one of several processing tasks based on the particular type of received data. Control information is supplied with the data to identify the data type and channel. The data sample is processed by the lower level task and execution returns to the main control loop to await more input data.

All processing of data is done at the time the sample is read from the input FIFO. There is no record of the past except for the running averages that are the display data. This means that all information contained in the sample is extracted when it is read and the sample is only processed once. The sample is averaged into a running total for the current pulse and a peak detection algorithm is applied. Peak detection compares the current sample and the previous sample. If the current sample marks a peak, the, pulse information is computed and the displays are updated.

An estimator control loop runs as a stand-alone piece of code with no inputs from the rest of the software system. It passes the estimator results to the background process and sampled Doppler data. Front panel switch and system information is sampled in the foreground, passed through the FIFO and processed in the same manner as Doppler data. If any change in the system environment has occurred, the appropriate information is updated in the control and data structures and reflected to the user.

Data structures allow related information to be grouped together and referenced as a unit. For example, all the data related to a Doppler processing channel can be grouped and referenced as "channel x". In a multi-channel environment, similar data types are processed identically, but at different times. The low level tasks can be written for a generic channel and passed via the identifier to the channel that is currently of interest. This reduces the amount of data passed from task to task since only the identifier has to be sent. All the channel data is then referenced with respect to a particular channel.

The monitor software uses two basic data structure types, one for channel information, and one for the current state of the world. The channel structure type is used twice for the left and right Doppler channels. It contains all information relevant to that channel, from status flags, to peak detection reference points, to current pulse and average flow data. The world data structure contains the current state of all hardware I/O ports including the displays and front panel switches.

These data structures provide an orderly means of transferring and updating information in both the software and hardware environment. When a new data sample is received, the low order tasks which are invoked process the sample and the appropriate channel's data set. If the sample ends a pulse period, the new pulse information is transferred into the world data structure and is passed to a display routine that transfers it to the front panel. Likewise, when a change is detected in a front panel switch, the new display operating mode is generated by pulling pertinent information from each channel's data set to update the world structure for transfer to the displays.

A special data structure is maintained for the speaker tone control due to a hardware requirement forcing tone generation into the foreground task level. By this arrangement, only tone information is made global to both task levels, thereby maintaining good isolation of interrupt handlers.

The data structures hold run-time information that periodically changes as blood flow is processed. All static information that defines the operating characteristics and processing parameters of the software system are defined as constants and must be declared at compile time. These include the sampling interval times, the averaging weight functions, and the scaling constants for the displaying values. These can be easily modified but require regeneration of code to take effect.

The following definitions are provided as reference to variable names and their functions in the module descriptions (algorithm Figures 24-41) and the pseudo-code (source code) associated with the module descriptions. For completeness of disclosure, all variable names and , source code employed by the software system are provided. However, it should be noted that not all of the definitions are referenced in the following monitor operation description, some are indicative of internal source code which is employed by the monitor's software, and some represent monitor hardware components, but whose particular identification is not essential to enable appreciation of the various module descriptions. CHAN_L left channel data structure identifier CHAN_R *right channel data structure identifier WORLD world data structure identifier TONE *tone data structure identifier CHANNEL: SAM_L *last data sample SAM_N *new data sample MIN *value of latest minimum point MAX *value of latest maximum point BASE_LINE *average velocity baseline

POINTS_PULSE number of points, last peak to current POINTS_MAX *number of points, last max to current POINTS_MIN *number of points, last min to current POINTS_TP *number of points, last trough to peak SUM_PULSE *sum of points, last peak to current SUM_MAX *sum of points, last max to current SUM_MIN *sum of points, last min to current CREST *peak to trough height PEAK channel peak height PULSE *current pulse value

AVE_PULSE average pulse rate AVE_VELOC *average velocity FST_MOM *average first moment INAF current INAF flow value AVE_INAF channel average INAF value PPE current PPE flow value POWER current Doppler power VOLUME current tone volume TONE current tone value X_SECT decode vessel diameter FLAGS set of channel status flags

SLOPE current veloc slope (l=up) DIR direction of search (l=peak)

BROKEN disconnected lead (l=broken lead) NO_PULSE no pulse detected (l=no pulse)

LO POWER Doppler power low (l=low power) CHAN channel identifier (l=left) TOG display status flag

PEAK_FND peak found WIN_SUM array of last N pulse sums WIN_CNT array of last N pulse point counts WIN INAF array of last N pulse INAFs

WORLD:

DISPLAYJ value in left display DISPLAY_R value in right display DECIMAL position of decimal point CONTROLS last version of port 1 controls VOLUME current volume FLAGS world status flags BROKE_L left broken status

BROKE_R right broken status

LEFT mode switch in 'left' mode

BOTH mode switch in 'both' mode

RIGHT mode switch in 'right' mode

INAF INAF data selected (0=PPE)

AUDIO audio select mode

BAT_ON battery indicator (l=on)

LOW_BAT low battery indicator

FLASH state of flashing displays (l=on)

FLASH_L left chan flash (l=flash enable)

FLASH R right chan flash (l=flash enable)

TONE:

NEW_TONE tone flag (l=new tone) FREQ new tone frequency WAIT CNT off-time counter

ESTIMATOR:

FREQ_NEW next position of estimator frequency FREQ_CUR current position of estimator FREQ_PREV last position of estimator POWER_CUR current power of signal POWER_AVE average power of signal SET_POINT ideal power ratio of peak frequency

OTHER DEFINITIONS: TIMER_1 80C196 internal timer #1

PORT_l 80C196 I/O port 1

CAM 80C196 event trigger memory

PWM 80C196 pulse width modulation register

HSI 80C196 high speed input ports HSO 80C196 high speed output ports

FIFO first in, first out memory buffer

PWR THRES threshold of "low power"

PWR_WGHT power averaging weight (25%)

VEL_WGHT velocity averaging weight (25%) FREQ_VELOC freq to velocity conversion factor

SAMP_PER channel sampling period (10 msec) INP_RATE input switch sampling rate (100 msec) PPE_TBL PPE conversion lookup table DIAM_TBL vessel diameter lookup table RATE_MAX maximum heart rate (300 beats/min) RATE_MIN minimum heart rate (30 beats/min) FLASH_CNT flash cvnter FLASH TIM display flashing time constant (1/4 sec) TONE_OFF tone off time (300 msec) LOOP GAIN loop gain variable

The operation of the monitor 2 will thus be appreciated by reference to Figures 24 through 41 in accompaniment with the following description thereof. The MAIN control loop (Figure 24) runs as the background task, scheduling events. It is invoked by a hardware reset to the microcontroller and is the first code executed after power-up. This only occurs if the power switch 10 is toggled or the battery dies. MAIN begins by executing the initialization task, INIT (Figure 25) followed by a polling loop that waits for input data from the input FIFO.

The INIT procedure is the first routine executed after power-up. It initializes the displays, the software environment, and all RAM based variables and data structures.

The digital displays 18 and 20 are initially turned off to avoid larger current surges while the power supply stabilizes and to prevent momentary garbled displays. Once the software system is initialized, a display test mode is entered which turns all segments on for 1/4 second. The test is performed on all other front panel indicators as well. The displays 18 and 20 are then set to dashes (" ") until normal processing produces flow information.

The stack pointer, the input FIFO pointers, and all other RAM based variables are initialized. Next the two channel data structures are initialized to indicate disconnected leads and zero flow with no pulse detected. The state-of-the-world data structure, WORLD, is initialized with a set of controls and flags to process both channels in "INAF" mode with the battery indicator off. The channel data is transferred into WORLD via two calls to the world update routine, WORLD UPDATE (Figure

26), and the display routine, DISPLAY (Figure 27), transfers the information to the front panel 6. The WORLD_UPDATE routine transfers display and output information from a channel data set to the current state-of-the-world (WORLD) data structure. It is invoked by routines such as INIT (Figure 25), PULSE (Figure 39) or SAMPLE (Figure 37) that change the current channel information or change operating modes. It is usually followed by a call to the DISPLAY routine (Figure 27) to transfer the information to the output devices).

The WORLD_UPDATE routine is passed a pointer to the channel structure of interest (CHAN_L or CHAN R) and the current world state (WORLD). It returns a display flag, either DISPLAY_L or DISPLAY_R, if new display data occurs. First a temporary variable is loaded with the appropriate display data based on the INAF/PPE status and broken lead flag. Next the mode switch separates LEFT and RIGHT setup mode from normal (BOTH) mode processing.

In setup mode, the left display gets the flow data and the right display gets the channel Doppler power level. The channel ID flag (CHAN) is tested against the LEFT or RIGHT switch status and the display is only updated if a match occurs. The tone is set to the appropriate frequency and volume. A broken lead disables the tone. In normal (BOTH) mode the appropriate display (left or right) is updated with the flow data and the tone disabled. Once the appropriate display flags associated with setup of the system are established for the left and/or right channels, * the WORLD_UPDATE routine is exited whereupon INIT (Figure 25) executes the DISPLAY routine.

The DISPLAY routine (Figure 27) provides an interface from the software environment to the hardware I/O devices. It is invoked whenever something in the WORLD data structure changes and transforms the status flag and display values to LED digits, lights and controls. The segment displays are updated with the new

DISPLAY L, DISPLAY R values. This requires translating the value from an 8 bit hexadecimal number to a three and one-half digit Code B number. A series of divides by 1000, 100, and 10 with each result representing a digit and the remainder being the new dividend provides the four digit values. These are written to the displays 18 and 20, and, if the battery is not "on", they are written to the "double digit" for extra brightness. The decimal point is set by the mode of operation.

The lights of the mode switches 28, 30 and 32 are driven by driving the PORT-1 I/O pins of the 80C196 microcontroller low if a switch is active and clearing the rest of the port pins. This lights the LED of the selected switch and locks the switch on. Next the volume is set by copying the VOLUME value to the PWM register.

The battery light 12 is turned on or off based on the BAT bit in the flag word. If the low battery flag

LOW_BAT is set and the flash enable bit is false, the light is turned off regardless of the BAT flag. The display flash control bits are disabled unless the channel flash enable bits, FLASH L or FLASH_R and flash enable bits are true.

Data sampling is initiated by starting an A/D conversion on A/D channel 0 and control is returned to the MAIN control loop (Figure 24) to await the first data in the input FIFO. By setting the initial state to "broken leads", the normal processing channels can be used to initialize the actual state of the world as soon as sampling begins. This simulates a case where a lead is disconnected and then reconnected, and avoids putting arbitrary data on the displays until actual flow information- is received. All computation variables for channel processing are initialized in the reconnect sequence.

Once the DISPLAY routine within INIT is executed, the INTERRUPT handler (Figure 28) is invoked by an A/D converter interrupt via the A/D interrupt vector and handles the foreground processes. These include controlling the estimator hardware, processing and passing velocity and power information to the FIFO and sampling the input switches and controls. Four routines are associated to these processes, ESTIMATE (Figure 30) which runes the estimator, INT_POWER (Figure 32) which processes Doppler power, and INT_SLOW (Figure 35) which performs switch/input processing.

When invoked, the A/D converter is read and one of four cases is processed based on which A/D channel interrupted. The sampling of A/D channels is set up as a chained sequence of conversions starting with A/D channel 0. Channel 0 convert is triggered by an event loaded into the internal CAM device and is precisely timed with respect to internal TIMER_1. When the conversion is complete, an interrupt occurs and channel 0 interrupt handler is called. The handler starts a conversion on A/D channel 1 and returns. This sequence continues for all four channels.

A/D channel 0 (case L_DOP) (Figure 29) has Left channel Doppler information, or the Left estimator filter power. This is sent to the ESTIMATE loop (Figure 30) to come up with the new estimated frequency point.

The ESTIMATE routine provides the control loop for the high pass filter in the frequency estimator circuit. The estimator works by maintaining a small percentage of the total received Doppler power in the band of the high pass filter. The filter's corner frequency is moved up and down based on the filter to total power ratio. Thus the corner tracks the upper edge of the received Doppler frequency spectrum.

The data is compared to a preset noise floor, if lower than the noise floor it is set to zero so as not to track extraneous noise. Otherwise, the noise floor is subtracted out for computation purposes. Next a loop gain variable, LOOP_GAIN, is computed as a function of a preset scaling constant LOOP_GAIN and the current estimator position. This linearizes the gain over the frequency range of the filter.

The filter position eventually is an 8-bit word (256 values) that is written to a DAC to set the corner frequency. In the estimator it is 100 times greater to allow increment and decrement values to be computed and accumulated without suffering round-off problems from integer math.

If the instantaneous total received power is below the noise floor (or zero), the filter position cannot be computed and the value is cut in half. If this condition persists, the position will decay to zero, creating a tail on the waveform rather than an abrupt cutoff when the power disappears.

Otherwise the filter position is computed by dividing the filter power by the total power to obtain the power percentage ratio. This ratio is subtracted by the desired power ratio (set-point) to create a signed difference quantity, plus or minus with respect to the set-point. The magnitude of this difference indicates current distance from the set-point and the sign identifies in which direction. This difference is multiplied by the loop gain and added to the current filter position in accordance with the following equation and is the new filter position that is written to the DACs in the estimator.

new position = ((filter/total) - set_point) loop_gain + position

An overrun and underrun test are done to avoid wraparound during large excursions of the filter. Finally a three point curve fit is done averaging the last point with the new point and the one prior to it and saving the result as the current point to write to the FIFO for peak processing.

The result is sent to the estimator (l_dacl and 2). Every Nth sample point, a fitted version of the frequency point is passed to the FIFO for peak processing in the background. This allows the estimator to sample faster than the 10ms background rate. Currently N=2. Finally the next A/D channel (channel 1) convert command is issued and we exit ESTIMATE loop of the interrupt handler.

A/D channel 1 (case L_PWR) (Figure 31) has the next left channel total power sample that is sent to the INT_PWR routine (Figure 32) which processes new power samples. Two power variables are maintained for each channel, current and average. The average power is averaged only during the active portion of the pulse and is passed on to the FIFO for further processing. The current power is used in the estimator and is an average of the last two samples.

In INT_PWR the data is first compared to the noise floor. If below noise floor, the current power is set to zero and the average power remains unchanged. A ti er is incremented to see how long the 'dead' time lasts. If sufficient time (greater than a pulse) has passed without any power then the average power decays to zero. If the power is above the noise floor, the timer is cleared and the data is averaged with average power.

The resulting processed power is used in the ESTIMATE routine (Figure 30) and every Mth sample is written to the FIFO (about 100 msec). The next A/D conversion is started on channel 2.

A/D channel 2 (case R_DOP) (Figure 33) is identical to the case R_DOP (A/D channel 0) except the data processed is the right channel estimator data. Once the A/D channel 2 convert is executed, the A/D channel 3 convert is started.

A/D channel 3 (case R_PWR) (Figure 34) processes the right channel power sample. In addition it runs a sample counter for Nth and Mth counting. The counter increments for each set of all four A/D channels being sampled. It then tests for the Pth sample which indicates time to read the input switches (every 100ms). On condition that COUNT = MOD N, the INT_SLOW routine is then called.

The INT_SLOW routine (Figure 35) generates tone data and reads the I/O ports for new status and control information sending the data to the FIFO.

Tones are generated by the interrupt routine to prevent loss of a sampling interval by a background task clearing the CAM. Tones are generated using locked events in the CAM. The only way to change tone is to clear out the entire CAM thus potentially losing a sample trigger. Thus tone changing occurs in the interrupt routine just prior to the new trigger being loaded. The tone data structure is made global because no parameter passing happens on interrupt calls. If a new tone is requested, NEW_T0NE true, the old tone (if any) is cleared and a timer is set to produce a short off period for beeping. When the "off time" counter expires, the new tone is loaded into the CAM and remains until the flag is set by a background task.

The input status and control bits are read from

I/O port 1. The bits that are active low are inverted so all bits read the same (1 = TRUE). The two broken lead bits are read from I/O port 2 and combined with the above bits and a code word to be sent to the FIFO.

Finally, A/D channel 0 convert is triggered by loading an event into the internal CAM device that is precisely timed with respect to internal TIMER_1. The next A/D channel 0 start-convert-time is computed from a time variable, TIME, plus the sampling interval. TIME is updated and loaded into the processor CAM to trigger the conversion when TIMER_1 equals TIME. This establishes a strict timing sequence between A/D channel 0 converts and the chain of interrupts follow.

Once the A/D conversions are completed, the INIT routine (Figure 25) is exited, whereupon the MAIN routine (Figure 24) queries whether the FIFO is empty. If the FIFO is empty, the MAIN routine queries whether a display flag is set. If yes, the DISPLAY routine is activated and the appropriate display is displayed on front panel 6. If no display flag is set, the DISPLAY routine is bypassed and the MAIN routine again queries whether the FIFO is empty. Assuming that the FIFO is not empty, the MAIN routine reads the FIFO to determine the classification of the data and how such data is to be handled. When new input data is received, it is passed off to an appropriate procedure for handling. An input word is a 16-bit integer. It includes a 10-bit data field and a 3-bit data-type select field from the source code definitions set forth hereinabove. Six types of input data are possible:

Case

0 Channel 1 (CHAN_L) Power Sample

1 Channel 2 (CHAN_R) Power Sample 2 Channel 1 (CHAN_L) Doppler Sample

3 Channel 2 (CHAN_R) Doppler Sample

4 Input Switches and Control Data (I_0)

5 Error Data

The following describes the various treatment of the input data which occurs in the foregoing Cases 0 through 5.

The POWER routine (Figure 36) is invoked from the MAIN procedure to process a sample of Doppler power information. The channel data structure and new sample are passed and the sample is averaged into the running average Doppler power for the channel.

The Doppler power is really the signal strength of the detected return signal from the transducer. It essentially represents the number of reflectors in the insonified area that produce a Doppler shift, (i.e. , those reflectors that are moving). Thus the information is useful as an indicator of probe positioning over the vessel or loss of flow. Unfortunately these two cases are difficult to differentiate because they have the same effect on the signal. The power signal has little or no informational content of how fast the blood is moving. The monitor 2 keeps a running average of the Doppler signal strength with a time constant of a few seconds. The time constant is set in software with a constant, PWR_WGHT. Since the analog hardware averages the signal strength over the short term, the input is only sampled slowly (100 msec) and these samples are averaged in software to produce the power indication.

The power is an arbitrary indicator of signal strength and has no absolute value. The indicator will be numbers from 0 to 10 in approximately 100 steps with zero indicating no flow (no Doppler signal) and 10 indicating maximum strength. Some logarithmic function may be added later if necessary for better low end resolution. A lower threshold can be set in software to indicate low power and to effect flashing of the display 18. As the reader will appreciate, the aforegoing description applies to both a Case 0 (CHAN_L) or Case 1 (CHAN_R) power data sample.

The SAMPLE procedure (Figure 37) processes the incoming Doppler data samples for a Doppler channel, either Case 2 (CHAN_L) or Case 3 (CHAN_R). It is invoked by the MAIN loop each time a new Doppler sample is read from the input FIFO and is passed the data sample and the appropriate channel pointer. Since there is no sample storage or history of data, all processing associated with the current data sample is performed including pulse detection and averaging. If a pulse is found the pulse is processed.

Initially the data sample is saved into the data structure as a new data sample in SAM_N and the previous value saved as a last data sample in SAM_L. Next the channel flags are tested for "broken lead" condition. If so, the sample has no meaning and processing is complete. Otherwise the PEAK detection algorithm (Figure 38) is called. The PEAK routine applies a peak detection algorithm to the new data sample and extracts the pulsatile information from the velocity waveform. It is called from the SAMPLE procedure and updates several variables and flags in the channel data structure as well as returning a peak found flag.

The peak detection algorithm uses a pair of data samples (SAM_N and SAM_L), a set of status flags (DIR, SLOPE) and two variables, MIN and MAX. The slope of the waveform is determined by comparing the new sample to the last one. The slope is declared "up" if the new point is greater than the old and "down" if the old point is greater than the new one. A change in slope is flagged when the new slope is different from the previously computed slope stored in SLOPE. If the slope has changed, a point of inflection has occurred and further processing is required. Otherwise the sample is just a continuation of the waveform in the same direction and the routine returns with no peak found.

Points of inflection are tagged as either peaks or troughs based on the slope which may or may not be valid pulse delimiters. The algorithm searches for a peak followed by a trough to validate a pulse. A peak must meet predefined criterion to be declared valid as must a trough. These criterion are tested in the VALIDATE function (not illustrated) and basically require that a sufficient deviation of the waveform has occurred before declaring a valid pulse.

The VALIDATE procedure takes a change of inflection point, either peak or trough, and decides whether that point has enough deviation to use it for pulse detection. The function operates on the channel data structure, CHAN, and returns a validation flag. To a first order of magnitude, pulse differentiation occurs when the signal deviates n% from some baseline value. A possible criterion is to average the peaks of each pulse and take a 10% deviation of this average as a sufficient change to detect pulsatile action. Other possible criteria using, for example, averaged first moment, or peak to trough difference, or the like, may be employed to determine pulse differentiation.

Several additional qualifiers can be applied to remove large noise glitches such as measuring peak to peak distances for reasonable spacing, and determining actual slope of the waveform. These are computed from the point counters POINTS_MAX and POINTS TP and the MIN and MAX variables. Troughs are validated in much the same manner.

Since a peak is defined by a following valid trough and vice-versa, either a valid peak or a valid trough at a time, but never both, needs to be found. A flag called DIR provides a log of what to look for, either valid peak or valid trough. Each time a valid trough is found, the direction variable, DIR, is switched to "peaks", and each time a valid peak is found, the direction becomes "troughs". Note that a valid condition is only determined some time after, the peak or trough actually occurred, thus it is possible to be at a new peak when the valid trough is detected. The MIN and MAX variables hold the sample value of the last valid trough or peak and are used as memories for comparisons of further points of interest.

To process a new point of inflection, the routine checks the DIR flag for "peak" or "trough" searching. If the search is for troughs and the point is a peak, it could be a qualifying point for the previous trough. The VALIDATE routine is called to test the deviation since the last MIN inflection point. If sufficient, the smallest of all previously detected troughs (which is stored in MIN) is declared a valid trough. The point we are at is high enough to be a valid peak so the sample value is stored in MAX and the direction of search is switched from "trough" to "peak".

If the current inflection point is a trough, and the search is for troughs then the new point may be lower than the previously detected lowest trough (in MIN). This is tested by comparing the current minimum point to MIN and saving the smaller of the two. In addition the accumulator that counts points from the last trough (POINTS_MIN) is reset. If the compare result is false, processing terminates.

The algorithm is fairly symmetrical in looking for troughs or peaks. Therefore, the processing of peaks is the same as above except that troughs become peaks and MINs become MAXs, etc. When a valid peak is decided, the peak found flag, PEAK_FND, is set and the no pulse flag, NO_PULSE in the channel data structure is cleared.

The PEAK routine returns a flag indicating whether a new valid peak (or pulse) has been found. If processing under a "no pulse" condition (i.e. constant nonpulsatile flow) a test is performed to inject a false pulse once a second for display update purposes. This test is done by checking the number of samples since the last pulse.

Next, the sample is accumulated into a running average of samples since the last pulse. The value is added to the variable SUM_PULSE and the number of sample points, POINTS_PULSE, is incremented. Since a peak is only validated sometime after the peak occurred, a second average, SUM MAX is maintained which was started at the last potential peak. By subtracting the second average from the first (when a peak is detected) the real average can be determined.

If a peak was detected (either real or false pulse), the PULSE routine (Figure 39) is invoked to compute the various parameters of the pulse and update the WORLD.

The PULSE procedure performs the flow computation and averaging function on the velocity data. It is called from the SAMPLE procedure (Figure 37) whenever a pulse is detected and is passed the channel to process. It updates the flow parameters and returns a display flag.

The SAMPLE routine keeps a running total of the samples since the last detected pulse. A pulse is defined as a trough-to-trough distance with a MAX point between. Since the trough is only detected at the peak following it, and the running total contains data to-date, samples from last trough to-date must be subtracted out. To achieve this objective, a secondary running total of points since the last trough is maintained (POINTS_MIN) . It is reset every time a minimum point is found and updated at each sample thereafter and is computed according to the following:

# points in pulse = (# points since last pulse) -

(# points since last min point).

Next, an averaging function is applied to the pulse reading to reduce display jitter. A simple timed average whose time constant is controlled by a single parameter is used. Flow is then computed by an approximation method using the peak velocity and the pulse rate to determine a time averaged flow for that pulse. The frequency sample is converted to velocity using the standard Doppler equation and then to flow by multiplying vessel cross sectional area with the velocity, where:

flow = velocity vessel diameter.

Thereafter, a windowing function is performed to average a "last N pulse" set of data for smoothing. This is done to both the idealized normalized absolute flow

(INAF) and pulse rate and is used in BOTH mode operation.

Two pulse averages are computed for setup modes.

The pulse palpitation equivalents (PPE) can be computed. The PPE number is derived from a set of comparisons with a table containing the upper and lower limits of each PPE range. Four PPE ranges are currently provided.

The display variables in the WORLD data structure are updated and the display flag set. If the average pulse rate exceeds 100 beats/min, only every second pulse is displayed. A flag is maintained that marks the even and odd pulses and the display flag is not set on the odd pulses.

The sample accumulators and counters are reset to start the next pulse and the SAMPLE routine processing completes. Under a "no peak found" condition, the running pulse count is tested for a time-out (less than 30 beats/ min) and the no pulse flag (NO_PULSE) is set if this is the case. The averaging accumulator and counter are cleared to indicate a start of pulse and operation resumes on the next sample under a "no pulse" condition. The I_0 routine (Figure 40) processes the user control and system status information Case 4, supra, and modifies the software and hardware environment appropriately. It is invoked by MAIN when a switch/ controls data word arrives from the front end sampling FIFO and modifies the WORLD data structure. It returns a display flag if anything has changed.

The routine starts by maintaining a counter to control the slow flash rate for displays. Since switch/control word sampling is synchronous and timely in the front end, it provides a good time base for other timers. Each time the routine is invoked, the flash timer, FLASH_CNT is incremented. When the timer reaches the preset interval, FLASH TIM, a flash control bit is toggled. The flash control bit, FLASH, indicates the on/off state of any display that should be flashing. In addition the routine tests the two channel flag words and sets a flag in WORLD for each display if flashing should be enabled (i.e. , no-pulse or low Doppler power).

The controls data word from the input FIFO is then compared to the flags word in WORLD. This word (CONTROLS), contains the last state of the input conditions. If anything has changed, a set of tests is performed on each bit field and, if necessary, the world status flags updated. For inputs that come from front panel switches, a deglitching control is performed. The switch has to be active for at least two sample states to be considered valid (100msec). The new input, the control word, and the flags provide the three deglitching states required.

The mode switch requires some special attention because the software provides the interlock between the switches. Only one of the three switches is active at a time and is marked in the flag word. If an new switch is pressed and deglitched, the software must release the switch that was set and set the new one. A test is also performed to arbitrarily pick one of two switches that may be pressed.

The broken lead test is performed on both channels and tests both disconnect and reconnect conditions. A reconnect causes the ENCODE function (Figure 4) to be invoked which encodes the transducer and resets all the channel variables.

The ENCODE routine (Figure 41) is invoked any time a "broken lead" has been repaired. It reads the transducer type, sets the diameter flag, and initializes the channel data structure for pulse detection and flow averaging. It is called from the I_0 routine on a change in the status of a channel's RF power bit. During power-up initializing, the "broken" status is forced "true" so that ENCODE is called on the first I/O sample.

Transducer encode is performed by a timing generator in the front end circuit whose time constant or frequency of operation is a function of an external resistor in the transducer element. By measuring the period of the returned waveform the resistive element in the connected transducer can be determined, thereby producing the appropriate vessel diameter constant for flow computations. Sixteen different value ranges are presently believed sufficient to cover most flow vessel diameters.

To measure the period, the HSI input feature of the processor is used. It precisely captures the state of TIMER_1 on every transition of the HSI input and loads it into an internal FIFO. The ENCODE routine clears out the FIFO and waits for two samples to arrive. This defines the period of the waveform and a series of compares assigns the appropriate number to the transducer. The vessel diameter is a predefined constant for each type of transducer or sensor 48. The channel data structure is initialized from defined start values.

Because most of the possible environment changes require information from both channels, the world update routine, WORLD_UP is invoked for both channels regardless of the change. The battery condition test is performed after the world update as no channel information is required if a change occurred.

Case 5 represents the situation wherein the detected input data is not properly classifiable into any of Cases 0 through 4). In this situation,, an unillustrated ERROR routine is called in which [DR. LAUB, PLEASE INSERT WHAT OCCURS UPON DETECTION OF AN ERROR].

The various procedures (Cases' 0 through 5) are invoked based on the select field and each returns a flag indicating whether a change in operating conditions or status has occurred. If so, the DISPLAY loop is invoked to update the hardware to reflect the change. Control then returns to the top of the data input loop of the MAIN routine awaiting the next input data word. The foregoing process continues as long as the clinician desires to monitor the patient.

Figure 42 represents a hardware arrangement, for executing the control and processing procedures described hereinabove in connection with Figures 24-41. Although only one processing channel is shown, i.e., for either the left or right channel, two such processing channels are provided in monitor 2.

As previously mentioned, an Intel 80C196 16-bit microcontroller, herein designated by reference numeral 102, preferably provides the processing power for the Doppler signal quantification and user interface. This chip is preferred because of its good development support and high level language, its 16-bit mathematical capabilities, high speed instruction rate, and low cost. In addition, it has a wide array of applicable on-chip functions, such as the A/D converter, PWM (pulse width modulated) output, internal timers and HSO (High Speed Output) function, that simplify the system design. It will be understood, however, that any suitable chip capable of performing comparably to an Intel 80C196 chip may be employed as the microprocessor or central processing (CPU) unit of the blood flow monitor 2 according to the present invention.

An 8 Mhz crystal provides the master reference clock to the processor 102. The resulting processor state time is 250 ns, with a 16-bit addition taking 1 μsec and instruction times averaging from 0.7 to 2.2 μsec. All circuit timing is designed and tested to run at a 10 Mhz master clock rate providing a potential 25% increase in performance if future software requires it.

Ports 3 and 4 of the microcontroller 102 are configured as the system bus 104. Port 3 serves as both the low order address (A0-A7) and 8 bit data'bus (AD0-AD7) and port 4 provides the high order address bus (A8-A15). A typical bus cycle puts the 16 bit address on the bus followed by an address strobe (ALE). ALE clocks the lower 8 bits into a holding latch for the rest of the cycle, the upper 8 address bits are held by the processor. The lower 8 bits are then driven with the data and a write strobe (WR-) for write cycles. Read cycles drive the read strobe (RD-) low to enable the peripheral device to drive the bus. All memory chip access cycles run with no wait states (4 system 'state' clocks). Both the RAM and EPROM chips 106 and 108 have access times selected to comply (150ns g 8MHz, 120ns §10 MHz). The displays 18 and 20 and DAC 110 writes require one wait state access, which adds 3 state times to the write cycle. This accomplished by driving the READY line low when display decode is detected. The number of wait states to insert is a programmable function in the processor and is nominally set to 1, thus READY can stay low as long as the address decode 112 is valid.

Both program and data memory are external to the processor, using conventional EPROM and static RAM chips, 108 and 106, respectively. The microcontroller 102 interfaces to the external memory via the multiplexed address/data system bus 104 on Ports 3 and 4. To reduce external memory component devices and to allow cost reduction, only 8-bit external memory access is allowed. An on-chip system bus controller converts 16-bit transfers to multiple 8-bit transfers, making the external bus width reduction transparent to the software programmer.

A 16-bit address bus allows 64K bytes of memory to be accessed. The external memory is decoded into three working areas, program EPROM, data RAM, and display/DAC interface. All address decoding is performed by a PAL device to allow flexibility in configuration. The following memory map gives the decode addresses. (All addresses are hexadecimal.)

EOOO-FFFF Unused 8 Kb

C000-DFFF Data RAM Static RAM

C000

B000-B016 Bar Graph DACS and

DISPLAYS A0OO-AO16 Estimator DACs 8000

9000-9016 Control register 32 Kb 8000-8016 Display drivers Program EPROM 0000-7FFF Program EPROM

0000

Up to 32 Kb of program is decodable. A 32Kx8 standard EPROM/PROM 108 (i.e. , 27C256 or equivalent) is preferably used for such decoding. The address decode for program memory is simply the most significant address bit (A15). Note that the first 100H bytes of code space are mapped internally to the microcontroller 102 and cannot be used.

The data memory space is configured to support up to 16 Kb of RAM. An 8Kx8, standard pin out static RAM 106 (i.e. , xx6264) or equivalent) is used in the presently preferred implementation. The static RAM is decoded into the entire upper 16K byte section of memory using A15. Preferably, only address C000H through the end of the inserted memory (C000+8K) should be used to avoid multiple decoding of the same physical memory. The region from 8000H to C000H is reserved for displays 18 and 20 and DAC 110 memory access. The first 16 bytes of data RAM are reserved.

The 16K byte block of address space from 8000H to C000H is used to decode the LED display driver 114 and front end digital to analog converters 110 (DACs) the bar graph divers 116 and a control register. The display drivers 114 are mapped into the lower 8K block and the DACs 110 into the upper 8K block. Each of the two display drivers 114 has 8 bytes of memory, so 16 bytes of display memory can be accessed. The right channel display maps into the first 8 bytes and the left channel into the next 8 bytes. The 16 bytes are decoded multiple times from 8000 to 9000H, although they should only be accessed in the first 16 bytes because to allow read-back capability of the display memory, the RAM space is decoded on top of display memory. Thus, a write to display memory also writes to the RAM at the same address, providing a copy of the display in the RAM.

To prevent RAM overwrites in display space the following rules are defined (all addresses are hexadecimal) :

1. Data writes from C000 to FFFF write to RAM chip only (although the first 16 bytes are reserved for display).

2. Data writes from 8000 to 800F write to both the RAM and the display memory. In RAM these are the 16 reserved bytes: C000 to C00F.

3. Data reads from C000 to FFFF read the RAM chip (where C000 to C00F is the read-back copy of the display memory in RAM) .

4. Data reads from 8000 to 800F also read the RAM copy of display data.

5. Addresses 8010 to 8FFF should not be used for either display or RAM access (even though it will work for both) to avoid false addressing.

A control register contains control bits for the programmable gain amps in the analog section and multiplexor control bits for the audio circuit. This register is mapped at address 9000H and acts as a single write only memory location. The PAL provides a positive- going write strobe. The control bits require a level translator to convert from TTL levels to the 12 volt control required for the multiplexors.

The DACs 110 are decoded into the A000H to C000H area. Each channel uses 3 DAC devices, two for the estimator filter 118 and one to drive the bar graph display 14. The estimator DACs are decoded in the lower portion of DAC space. Two write strobes, L_DACWR- and R_DACWR- are generated in the PAL using A3 to select left and right channels. Address bit A0 is used to select DAC 1 and 2 for each channel of the monitor. The bar graph drivers 116 are selected in the upper section of DAC space using A3 to select left and right drivers. The DACWR- strobe from the PAL strobes these devices.

Some of the specialized microprocessor ports are used for other system functions. The Port 0 I/O pins are configured to act as analog inputs to a 10-bit on-chip A/D converter. The analog estimator signal and the Doppler power indicators from both channels are brought to this port. Software driven control selects which pin to internally multiplex to the A/D converter and triggers a conversion. An interrupt occurs following the conversion (31 μsec). The sampling interval is set and triggered by an internal timer (TIMER-1) in conjunction with the HSO unit and CAM. Basically, software control presets an event (in this case an A/D conversion) to happen at a specific time with reference to a timer (TIMER-1). When the timer matches the preset event, the event is triggered. Thus precise timing intervals can be programmed via software and triggered in hardware, leaving the processor free to do background processing.

Port 1 is used as an input/output port for switch and status inputs. The switch function inputs are LEFT, BOTH and RIGHT select, INAP/PPE select, and AUDIO select. These come from the aforementioned front panel switches and are debounced in software as follows. The port pins are sampled every 0.1 seconds (approximately) and compared to the last value: if a pin changes state and remains changed for two samples, the new mode of operation is entered. The sampling interval is set so that events will not be missed. The battery and power status are monitored in the same manner.

The MODE switch interlock is controlled by port 1. The port pins are bidirectional I/O lines that can be driven by the processor 102. In the high state, the output resembles an open collector with a weak pull-up resistor. Normally the switch pins are driven high. When a switch is depressed, it pulls the pin low momentarily. The processor detects this and drives that pin low and releases any previously driven pin. Thus the processor latches the switch and provides an interlock between the three mode switches.

One of the High Speed Output (HSO) pins, in conjunction with internal TIMER-2, is used as the source of the computer simulated Doppler audio tone. A pair of events are locked into the HSO unit to toggle the HSO output pin (HS03) at a fixed frequency (f=l/T). One event sets HS03 "ON" at time T/2 and the next event clears HS03 at time T and resets TIMER-2. This results in a square wave output at frequency 1/T, where T can be programmed in

1.6 μsec steps from 10Hz to 312K HZ (approximately).

Desirably, an audio range of about 200 Hz to 5KHz is chosen for the simulated Doppler tone. Software routines break the tone into 'beeps' that are proportional in length to the pulse rate.

Five other HSO outputs are used. HSO0 and HSOl provide low frequency flashing strobes for the two LED displays and HS02 directly drives the battery LED. Since the blink rate for these LEDs is slow (.5 to 1 Hz) the internal timers cannot be directly used. A software timer is used to provide the slow flashing signal. The remaining two are multiplexor controls. BEEP sets the audio mux for electronic tone output and CAL allows an artificial signal to be injected into the estimator for calibration.

The transducer encoding function is performed using the HSI inputs. These pins cause the state of internal Timer 1 to be loaded into a FIFO when a transition occurs. Precise edge timing can be performed in this manner. The transducer encoding function produces a timed waveform whose period is a function of the transducer in use. The period is measured by the computer and the vessel diameter found using a table lookup.

An 8 digit LED driver circuit 114 provides an interface between the microprocessor and the LED display. A separate driver chip is provided for each three and one-half digit display. This allows the left and right displays to independently flash and allows a 1/2 brightness mode to be enabled when running on batteries.

The controller/driver 114 receives data from the processor, stores it, decodes it to 7 segment format and multiplexes one of 8 digits onto the LED driver bus 104. Thus, only one display digit is decoded and driven at a time. This reduces the number of wires/pins to the display, decreases the total LED drive current and increases display efficiency by driving the LEDs at higher current (for a short time) where the luminous output per unit current is high. The multiplex scan rate is around 250 Hz so no flicker is seen.

The LED brightness is a function of digit on- time and forward current. The presently preferred configuration uses a common anode driver with the digit driver supplying all segments with forward current and the segment outputs sinking the individual segment current. The forward current of each LED is limited by a series resistor, resulting in approximately 10 mA current. The LED displays chosen are low current displays with rated luminance output at 1 mA per segment. Since each digit is activated for approximately 1/10 of the time, (1/8 cycle plus some off-time between digits) a peak current of 10 mA gives an average current of about 1 mA. Since each driver is only driving 4 of the 8 possible digits, the brightness of the display can be doubled simply by connecting 2 of the 8 digit drivers to each digit anode and turning it on for two of the digit time intervals instead of one. The software supports this function by programming the same value in the two memory slots assigned to a digit, or leaving one of the slots blank (1/ 2 brightness). The digits are connected so that the lower 4 digits drive the 4 LED display units and the upper 4 digit addresses double the respective brightness.

Figure imgf000060_0001

The processor 102 writes to the display controller 114 by providing a 3 bit address to select one of 8 digit memory slots, plus a 4-bit data word defining the character to display and a write strobe. An additional data bit accompanies the data word to enable/disable the decimal point associated to that digit. A normal one wait-state memory cycle from the processor is adequate for the selected display driver, although a latch is provided on the main digital board to increase data hold time for a second (cheaper) source. The latch also provides buffering for the microprocessor bus.

The 4-bit processor data is decoded into a code

B format which gives the display characters:

data: 0 1 2 3 4 5 6 7 8 9 A B C D E F

I I I I I I I I I I I I I I I I char: 0 1 2 3 4 5 6 7 8 9 - E H L P (blank)

The 1/2 digit that displays the +/-1 function requires special encoding and is wired to give the following:

data: 0 1 2 3 4 5 6 7 8 9 A B C D E F

I I I I I I I I I I I I I I I I char: + 1 1 ?-l ? ? ? ? ? - ? ? + ? (blank)

(where ? = unknown)

The displays are " flashed by turning of the segment drivers for a period of time using a shutdown pin. This is a three function pin where the undriven state puts the chip in code B mode and the low state puts the chip in shutdown mode. An open collector driver to this pin allows simple strobing from the microprocessor, where HIGH means normal operation and LOW blanks the display 18 or 20. The internal memory remains unchanged during shutdown or can be updated. A separate control is provided for left and right displays.

The bar graph displays 14 and 16 are driven from a bar display driver chip 116 that converts an analog input to lighted bars. The analog input is driven from a pair of DACs 110 on the main PC board which simulate the velocity signal from each channel's estimator circuit. The chip 116 uses a logarithmic scale in determining the output. The individual LED current is set by a single reference (using two resistors) that controls a constant current 'sink'. Low current (1 ma) LED bar graph displays 14 and 16 are preferred to reduce power consumption.

The analog front end of the monitor 2 performs the standard Doppler experiment. A transmit circuit drives the transmit crystal or oscillator (OSC) 120, insonifying the patient. A receive circuit, including mixer demodulator 122, detects the Doppler shifted return signal from a receive crystal and converts it to a base-banded Doppler signal for audio output and quantization. The quantization is done by the estimator circuit in conjunction with the microprocessor 102. In addition, an encode circuit 124 provides transducer encoding.

A transmitter of the microprocessor 102 preferably drives the transmit crystal 120 at the ultrasonic frequency of 9.1 Mhz. The transmitter is designed to drive a 50 ohm transducer/sensor 48 with minimal intertransducer variation. The transmitter drive circuit operates in class C mode to provide maximum efficiency. A MOSFET provides the current switching into the tuned load (tank circuit). A step-down transformer provides isolation to the patient as well as the load matching function and a tuned class C amplifier 126 is capable of driving twice the supply voltage. The coupling transformer has a turns ratio of three and is used as a step down coupler. The circuit is configured to drive up to 10 volts pp into the 50 ohm "transducer assembly.

Within the microprocessor, a 9.1 Mhz oscillator circuit provides the transistor base drive to ensure clean (unmodulated) drive. The amplifier 126 is driven directly with the square wave output for maximum power. The circuit configuration ensures that the maximum safe transmit levels can never be exceeded.

The encode circuit 124 produces a means by which various transducer elements attached to the lead set can be identified. Each disposable transducer contains a programming element (a resistor 84, Figure 9) identifying the device. An extra set of leads connects this element to the monitor where it can be interrogated by the microprocessor to set various processing parameters.

Among the objectives of the instant invention are to provide a transducer programming element of minimal cost, to allow a plurality (e.g. , 16) different sensor elements to be accurately identified, and to maintain isolation of the transducer lead set from the instrument circuitry. These goals are achieved by a timing circuit that uses the resistive programming element to set the timing interval. Timing pulses are passed to the microprocessor 102 which reads them using its HSI inputs for accurate event timing. The period of the received waveform is measured, producing a lookup for the appropriate lumen diameter parameter set.

The timing circuit which includes oscillator 120 draws its power from an RF signal on the secondary of the transmit transformer of the microprocessor. The RF signal is rectified and filtered to produce a DC supply voltage. The received timing pulses are transmitted to the microprocessor via an optically isolated device. Information is represented by pulse repetition rate, which are economically passed across the patient isolation barrier, i.e., with no distortion. The circuit operates at very low power and is highly accurate.

A CMOS 555 timer is used as the timing generator. It runs in an astable mode and produces a periodic pulse train proportional to the timing resistor element and a fixed capacitor. A plurality (e.g. , 16) resistor values are selected to ensure sufficient spacing between the "codes" for reliable detection. To reduce parts cost, the timing element tolerances are somewhat liberal so that the code spacings must be fairly wide. The foregoing, in conjunction with a maximum timing resolution of the micro HSI inputs sets the range of timing periods (i.e. , encoding resistors 84).

The Doppler receiver circuit converts the RF signal from the receive crystal 72 of the transducer or sensor 48 to a basebanded audio signal containing the Doppler shift frequencies.

The returned signal contains ultrasonic energy from various sources. The signal of interest is the

Doppler shifted energy from scatterers (red blood cells) in the moving blood. This return is very small and is cluttered by backscattered energy from surrounding tissue, as well as directly coupled energy from the transmit crystal 72 of the sensor 48. The Doppler signal is typically 30 to 80 dB lower than competing signals from the other sources. The receive signal is a superposition of all the ultrasonic energy signals picked up by the receive crystal, resulting in a carrier waveform comprised of the large stationary returns as modulated by the small doppler shifted returns. The Doppler shift frequencies are in the audio range and the signal therefore appears very much like an FM radio signal.

The basic receiver concept is to detect the audio signal by mixing (multiplying) the received input with a reference oscillator at mixer demodulator 122 at the carrier frequency to produce the sum and difference frequencies. A low pass filter 128 removes the sum frequency. The resulting difference frequency is the modulated carrier minus the carrier, or just the modulating (audio Doppler) signal.

The microprocessor 102 desirably uses a balanced demodulator circuit. The input signal is transformer coupled and passed directly to the balanced demodulator. The demodulator has a dynamic range in excess of 80db and can directly detect the Doppler shifted information. The chip output contains the detected Doppler audio signal. Single pole filters remove the sum frequency component and the signal is amplified by a gain stage of approximately 1000.

The audio output is band pass filtered using several single pole high and low pass filters at the various gain stages. The low frequency corner is set to remove noise generated by slowly moving tissue and vessel walls. The high frequency corner is set at the maximum expected frequency and removes any process induced noise and high frequency interference. The bandpass is set from 200 Hz to 7 KHz. This allows blood flow from about .02 to 1 m/sec.to be measured.

The various scatterers (red blood cells) in the insonified vessel move at different speeds, producing range or spectrum of frequency components contained in the Doppler audio signal. The estimator circuit determines the peak frequency component of this spectrum which is used to compute peak velocity and flow.

In the ideal parabolic flow model, the power spectrum of the Doppler signal normally ranges from 0 Hz to the frequency of the maximum or peak velocity component of the blood. The peak frequency is then the uppermost edge of the power spectrum envelope. In reality, this upper edge is somewhat distorted and spread out due to transit time effects, phase nose and instrumentation noise. The peak velocity point must then be found somewhere along the sloped high frequency edge of the power spectrum envelope. Under poor signal to noise conditions this point becomes more difficult to identify.

Since it is inefficient and costly to perform complete spectral analysis simply to detect peak frequency, the monitor estimator uses a more direct approach to finding and tracking the peak frequency waveform. A high pass filter in a feedback loop tracks the upper edge of the power spectrum envelope. Specifically, the central element is the high pass filter 118 which has a controllable corner frequency. The power in the filter pass-band is maintained at some small percentage of the total power in the Doppler signal. If the power ratio decreases, the corner frequency is lowered and if it increases, the corner frequency is raised. The ratio is chosen so that the filter corner intercepts the trailing edge of the power envelope at the point of the maximum velocity component. Thus the filter corner frequency tracks the maximum or peak velocity waveform.

The circuit is implemented using both analog and digital techniques. The high pass filter 118 combines a pair of digital to analog converters (DACs) with a conventional two pole active filter. The total signal power 130 and filter power 132 are prepared as analog signals that are sampled by the microprocessor 102 which in turn computes the ratio of powers and moves the filter corner frequency appropriately.

The microprocessor is included in the feedback loop for several reasons. Since the total power 130 is necessarily computed and sampled for display purposes, there is no additional demand or inconvenience required in obtaining this parameter. More importantly, the microprocessor enables the envelope intercept point to be precisely identified as a function of the current peak frequency and total power. This is especially helpful in low power situations where the signal to noise ratio is poor or the signal has been reduced to below the wall filter.

The analog power measurement is preferably computed using an RMS to DC conversion chip. This circuit finds the absolute value of the input signal as well as squares, averages, and takes the square-root of the signal. Power is found by squaring the result in the microprocessor, although this is not necessary since only the ratio is of interest. To measure the RMS value with any accuracy, the averaging time constant must be long enough for the lowest frequency component. The peak frequency estimating filter needs quick RMS power averaging sine the control loop time constant is fast (5-10 ms) .

The computer samples the filter power 132 at approximately 5 ms intervals. As mentioned previously, total power 130 is sampled with the filter power. The ratio of filter to total power is computed and an adjustment is made to the corner frequency. The direction of the adjustment and the amount to adjust are based on past history of adjustments and how close the ratio is to the desired point. This new adjustment is written to the DACs as the new corner frequency. Preferably, the filter can be set in 16 Hz intervals from 16 Hz to 4 KHz (256 steps) .

Since the loop time constant is about the same rate as he sampling rate the filter will continuously search for the ideal corner and will trail the actual peak waveform. Since the time constant of the circuit is known, the adjustments can anticipate a delayed response to a new corner setting. The algorithm is tuned in software for optimal performance. The corner frequency settings on each DAC represent the points of the computed peak velocity waveform. A curve fitting function performs smoothing by fitting each point between its adjacent neighbors using averaging methods. This removes noise spikes and errors induced by the power circuits.

The audio circuit provides the enunciator function of the monitor 2. Two types of audio output can be selected. As provided hereinabove, the raw detected Doppler frequency from either channel may be selected or a computer simulated version of the Doppler signal. The simulated output consists of a series of tones or beeps with frequency proportional to the computed flow. The beep duration is proportional to the length of a heart beat.

A quad analog switch performs the multiplexing function for the audio selection and a pair of analog switches enable one of the raw Doppler channels. The analog switches are selected by the signals L_AUDIO and R_ AUDIO from the control register. A third switch enables the computer generated TONE signal to pass through and is enabled by BEEP from HS04 on the microprocessor. The control states are a function of the input switches and are set in software. All control signals are passed through level shifting devices to convert the TTL signal to +12. The analog switch of the multiplexor preferably has an 'on' resistance of about 100 ohms, which is dominated by the 10K ohm coupling resistor to the next stage. Both input sources are fairly low impedance and the signal level is 0 to 1 volt. The computer generated TONE is a square wave with a frequency generally in the range of 300 Hz to 3KHz. A filter/attenuator network removes some high order harmonics and reduces the signal level to approximately 1 volt. The signal is fed through a divider network to reduce the signal level to about 100 mV. The lower leg of this divider is the volume control potentiometer knob 34 on the front panel 6. The volume potentiometer is a log wound potentiometer to produce a normal range of volume control. The resulting audio signal is fed to an audio amp 134 to drive the speaker 40 output. The audio amp has a gain of 20 and can directly drive an 8 ohm speaker. A switch in the earphone jack 26 disconnects the speaker when the earphone is plugged in.

The power supply circuit 42 generates the +5 volt digital supply, the +12 volt analog supply, provides uninterruptable battery backup, performs the automatic shutdown function and provides medical isolation for the instrument. A commercially available 12 volt medically isolated supply performs the AC to DC conversion and primary power source. The output is used as a constant voltage charger for a six cell NICAD battery pack 44 (3.6 volts). A pair of diodes act as switches to direct either the charging supply or the batteries to the regulation stage.

The analog supply uses a distributed power regulation scheme to optimize noise performance. A switching supply provides the nominal voltage which is then linearly regulated locally for sensitive circuit areas. The nominal 12 volt regulator circuit includes a step-up switching type that uses pulse gated control. An external inductor and switching MOSFET are designed to supply 15V at .5 amps with an input voltage range of 6 to 11.5 volts. The output is filtered with a 1000 μF capacitor and passed through linear regulator to obtain a clean 12 volt supply. Further regulation is provided for each channel's mixer and gain circuits and for the audio stage. The result is inexpensive and high quality power supply noise isolation. The regulators are 10 volt, low current, low overhead linear ICs.

The 5 volt digital supply is linearly regulated from the charger/battery supply 44. A low overhead regulator is required to ensure operation during low battery conditions of 6 volts. The regulator can handle one amp continuous draw if proper heatsinking is provided. A power resistor is used when running of the charger to reduce regulator device dissipation.

The automatic shutdown circuit monitors the battery voltage and shuts the unit off when the battery goes dead. A comparator circuit drives a relay that is closed anytime power is at a sufficient level. A self latching feedback diode ensures that the circuit remains off once triggered. Some additional filtering is desirably provided a the comparator input to prevent short load transients from tripping the circuit. And, it is further preferred that the circuit be fused internally with a 1.5 amp slow blow fuse. Maximum transient current is preferably 800ma and nominal current is around 500ma.

The front panel 6 of the monitor 2 is backed by an unillustrated circuit card that carries the displays and switches 10, 12, . . ., 36 and 38. Some of the display control hardware resides on this board to reduce the number of connections to the main digital board. The two cards are suitably connected by a 50 pin ribbon cable connector that passes all display and switch information, plus power and ground, for the display circuits. For the most part, the display board just provides a connection mechanism to the digital card. The switch pull-up resistors are mounted on the digital board to reduce noise at the processor pin.

Although the invention has been described in detail for the purpose of illustration, it is to be understood that such detail is solely for that purpose and that variations can be made therein by those skilled in the art without departing from the spirit and scope of the invention except as it may be limited by the claims.

Claims

CLAIMSWhat is claimed is:
1. A sensor device comprising a substrate including means for selected electrical conductivity and at least one signal means in electrical communication with said means for electrical conductivity.
2. A sensor device of Claim 1 wherein said signal means comprises a transducer.
3. A sensor device for use with monitoring equipment, said device comprising:
a substrate including means provided thereon for conducting electricity;
at least one transducer element carried by said substrate and electrically communicable with said means for conducting electricity; and
means for securing said substrate to a surface.
4. The device of Claim 1 or 3 wherein said substrate is substantially rigid.
5. The device of Claim 4 wherein said substrate is formed of ceramic material.
6. The device of Claim 4 wherein said substrate is formed of substantially rigid plastic. ,
7. The device of Claim 1 or 3 wherein said substrate is substantially flexible.
8. The device of Claim 7 wherein said substrate is formed of paper products.
9. The device of Claim 7 wherein said substrate is formed of substantially flexible plastic.
10. The device of Claim 1 or 3 wherein said means for conducting electricity comprise encoding means for enabling said monitoring equipment to identify operational characteristics of said sensor device.
11. The device of Claim 10 wherein said encoding means comprise a resistor.
12. The device of Claim 2 or 3 wherein said at least one transducer element comprises a plurality of transducer elements.
13. The device of Claim 12 wherein said plurality of transducer elements consists of two transducer elements.
14. The device of Claim 13 wherein said transducer elements lie in substantially coplanar relationship.
15. The device of Claim 13 further comprising means for disposing said transducer elements into substantially non-coplanar relationship.
16. The device of Claim 15 wherein said means for disposing comprise a shim situated between at least one of said transducer elements and said substrate.
17. The device of Claim 16 wherein said means for disposing further comprise a shim situated between each of said transducer elements and said substrate.
18. The device of Claim 15 wherein said means for disposing comprise a notch provided in said substrate and accommodating one of said two transducer elements.
19. The device of Claim 18 wherein said means for disposing further comprise a pair of notches provided in said substrate such that one of each of said pair of notches accommodates one of said two transducer elements.
20. The device of Claim 19 wherein said pair of notches are disposed in substantially reverse image angular inclination with respect to one another.
21. The device of Claim 19 wherein said pair of notches are disposed in substantially similar angular inclination with respect to one another.
22. The device of Claim 2 or 3 further comprising means for disposing said at least one transducer element into an angular inclination with respect to said substrate.
23. The device of Claim 22 wherein said means for disposing comprise' a notch provided in said substrate and accommodating said at least one transducer element.
24. The device of Claim 2 or 3 -wherein said at least one transducer element comprises a piezoelectric crystal having conductive material on opposite surfaces thereof.
25. The device of Claim 24 wherein electrically conductive adhesive secures said at least one transducer to said substrate.
26. The device of Claim 24 wherein said means for conducting electricity comprise through-bores provided in said substrate for enabling said electrically conductive adhesive to electrically connect said at least one transducer element with electrical trace elements provided on a side of said substrate opposite to said at least one transducer element.
27. The device of Claim 1 or 3 further comprising means for detachably connecting said device to a cable associated with said monitoring equipment.
28. A blood flow monitoring system comprising:
(a) a user-interactive monitor for generating electrical output signals corresponding to a first acoustic frequency, for processing electrical input signals corresponding to a second acoustic frequency and for displaying blood flow characteristics of a blood vessel being monitored;
(b) at least one sensor means comprising:
a substrate including means provided thereon for conducting electricity; and
at least one transducer element carried by said substrate and electrically communicable with said means for conducting electricity, said at least one transducer element being operable to convert said output signals to said first acoustic frequency and to transmit said converted output signals to a patient's blood vessel, said at least one transducer element being further operable to receive signals at said second acoustic frequency reflected from said blood vessel and to convert said received signals into said input signals; and means for securing said substrate to a patient's skin surface; and
(c) means for electrically connecting said monitor to said at least one sensor means.
29. The system of Claim 28 wherein said output signals are transmitted and said input signals are received in a continuous wave.
30. The system of Claim 28 wherein said output signals are transmitted and said input signals are received in pulses.
31. The system of Claim 28 wherein said output signals are transmitted and said input signals are received in phased array.
32. The system of Claim 28 wherein said at least one sensor means comprise two sensor means whereby two blood vessels of said patient may be simultaneously monitored.
33. The system of Claim 28 wherein said monitor includes means for determining said blood flow characteristics according to an idealized normalized absolute flow scale.
34. The system of Claim 28 wherein said monitor includes means for determining said blood flow characteristics according to a pulse palpitation equivalents scale.
35. The system of Claim 28 wherein said monitor includes means for determining signal strength of said input signals.
36. The system of Claim 28 wherein said monitor includes means for producing an audible signal corresponding to received signal strength of said input signals.
37. The system of Claim 28 wherein said substrate is substantially rigid.
38. The system of Claim 37 wherein said substrate is formed of ceramic material.
39. The system of Claim 37 wherein said substrate is formed of substantially rigid plastic.
40. The system of Claim 28 wherein said substrate is substantially flexible.
41. The system of Claim 40 wherein said substrate is formed of paper products.
42. The system of Claim 40 wherein said substrate is formed of substantially flexible plastic.
43. The system of Claim 28 wherein said means for conducting electricity comprise encoding means for enabling said monitor to identify operational characteristics of said sensor device.
44. The system of Claim 43 wherein said encoding means comprise a resistor.
45. The system of Claim 28 wherein said at least one transducer element comprises of plurality of transducer elements.
46. The system of Claim 45 wherein said plurality of transducer elements consists of two transducer elements, whereby one of said transducer elements transmits said output signals and the other of said transducer elements receives said input signals.
47. The system of Claim 46 wherein said transducer elements lie in substantially coplanar relationship.
48. The system of Claim 46 further comprising means for disposing said transducer elements into substantially non-coplanar relationship.
49. The system of Claim 48 wherein said means for disposing comprise a shim situated between at least one of said transducer elements and said substrate.
50. The system of Claim 49 wherein said means for disposing further comprise a shim situated between each of said transducer elements and said substrate.
51. The system of Claim 48 wherein said means for disposing comprise a notch provided in said substrate and accommodating one of said two transducer elements.
52. The system of Claim 51 wherein said means for disposing further comprise a pair of notches provided in said substrate such that one of each of said pair of notches accommodates one of said two transducer elements.
53. The system of Claim 52 wherein said pair of notches are disposed in substantially reverse image angular inclination with respect to one another.
54. The system of Claim 52 wherein said pair of notches are disposed in substantially similar angular inclination with respect to one another.
55. The system of Claim 28 further comprising means for disposing said at least one transducer element into an angular inclination with respect to said substrate.
56. The system of Claim 55 wherein said means for disposing comprise a notch provided in said substrate and accommodating said at least one transducer element.
57. The system of Claim 28 wherein said at least one transducer element comprises a piezoelectric crystal having conductive material on opposite surfaces thereof.
58. The system of Claim 57 wherein electrically conductive adhesive secures said at least one transducer to said substrate.
59. The system of Claim 57 wherein said means for conducting electricity comprise through-bores provided in said substrate for enabling said electrically conductive adhesive to electrically connect said at least one transducer element with electrical trace elements provided on a side of said substrate opposite to said at least one transducer element.
60. The system of Claim 28 further comprising means for detachably connecting said at least one sensor means to said means for electrically connecting said monitor to said at least one sensor means.
PCT/US1993/008105 1992-08-25 1993-08-25 Blood flow monitoring system WO1994004073A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US93552492 true 1992-08-25 1992-08-25
US07/935,524 1992-08-25

Publications (1)

Publication Number Publication Date
WO1994004073A1 true true WO1994004073A1 (en) 1994-03-03

Family

ID=25467298

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1993/008105 WO1994004073A1 (en) 1992-08-25 1993-08-25 Blood flow monitoring system

Country Status (1)

Country Link
WO (1) WO1994004073A1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6575914B2 (en) 2001-05-18 2003-06-10 Koninklijke Philips Electronics N.V. Integrated cardiac resuscitation system with ability to detect perfusion
CN103239238A (en) * 2012-02-07 2013-08-14 北京超思电子技术股份有限公司 Pulse blood oxygen measuring instrument
WO2014048911A1 (en) 2012-09-28 2014-04-03 Biofluidix Gmbh Capacitive pressure sensor

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4576178A (en) * 1983-03-28 1986-03-18 David Johnson Audio signal generator
US4621643A (en) * 1982-09-02 1986-11-11 Nellcor Incorporated Calibrated optical oximeter probe
US4685469A (en) * 1984-03-09 1987-08-11 Keller Hans W Piezoresistive pressure measuring cell
US4867557A (en) * 1987-04-09 1989-09-19 Sumitomo Electric Industries, Ltd. Reflection type oximeter for applying light pulses to a body tissue to measure oxygen saturation
US4901729A (en) * 1987-03-10 1990-02-20 Matsushita Electric Industrial Co., Ltd. Ultrasonic probe having ultrasonic propagation medium
US5024226A (en) * 1989-08-17 1991-06-18 Critikon, Inc. Epidural oxygen sensor
US5107840A (en) * 1990-05-22 1992-04-28 U.S. Philips Corporation Device for measuring and displaying of ultrasonic echography the flow rate of a blood flow and the dilation of the associated vessel
US5170786A (en) * 1990-09-28 1992-12-15 Novametrix Medical Systems, Inc. Reusable probe system
US5188108A (en) * 1990-02-15 1993-02-23 Hewlett-Packard Company Sensor, apparatus and method for non-invasive measurement of oxygen saturation
US5212387A (en) * 1992-01-27 1993-05-18 Charles H. Swan & Louis S. Pavloff, D.D. Ltd. Laser radiation barrier
US5217013A (en) * 1983-10-14 1993-06-08 Somanetics Corporation Patient sensor for optical cerebral oximeter and the like

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4621643A (en) * 1982-09-02 1986-11-11 Nellcor Incorporated Calibrated optical oximeter probe
US4576178A (en) * 1983-03-28 1986-03-18 David Johnson Audio signal generator
US5217013A (en) * 1983-10-14 1993-06-08 Somanetics Corporation Patient sensor for optical cerebral oximeter and the like
US4685469A (en) * 1984-03-09 1987-08-11 Keller Hans W Piezoresistive pressure measuring cell
US4901729A (en) * 1987-03-10 1990-02-20 Matsushita Electric Industrial Co., Ltd. Ultrasonic probe having ultrasonic propagation medium
US4867557A (en) * 1987-04-09 1989-09-19 Sumitomo Electric Industries, Ltd. Reflection type oximeter for applying light pulses to a body tissue to measure oxygen saturation
US5024226A (en) * 1989-08-17 1991-06-18 Critikon, Inc. Epidural oxygen sensor
US5188108A (en) * 1990-02-15 1993-02-23 Hewlett-Packard Company Sensor, apparatus and method for non-invasive measurement of oxygen saturation
US5107840A (en) * 1990-05-22 1992-04-28 U.S. Philips Corporation Device for measuring and displaying of ultrasonic echography the flow rate of a blood flow and the dilation of the associated vessel
US5170786A (en) * 1990-09-28 1992-12-15 Novametrix Medical Systems, Inc. Reusable probe system
US5212387A (en) * 1992-01-27 1993-05-18 Charles H. Swan & Louis S. Pavloff, D.D. Ltd. Laser radiation barrier

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6575914B2 (en) 2001-05-18 2003-06-10 Koninklijke Philips Electronics N.V. Integrated cardiac resuscitation system with ability to detect perfusion
CN103239238A (en) * 2012-02-07 2013-08-14 北京超思电子技术股份有限公司 Pulse blood oxygen measuring instrument
WO2014048911A1 (en) 2012-09-28 2014-04-03 Biofluidix Gmbh Capacitive pressure sensor

Similar Documents

Publication Publication Date Title
US3633569A (en) Arrhythmia counter
US6594511B2 (en) Method and apparatus for determining physiological characteristics
US6863652B2 (en) Power conserving adaptive control system for generating signal in portable medical devices
US5402794A (en) Method and apparatus for heart transplant monitoring and analog telemetry calibration
US6198951B1 (en) Reflection photodetector and biological information measuring instrument
US5462051A (en) Medical communication system
US4088125A (en) Method and apparatus for monitoring skin potential response
Labovitz et al. The effects of sampling site on the two-dimensional echo-Doppler determination of cardiac output
US4974607A (en) System for centralized management of medical data
US4890619A (en) System for the measurement of the content of a gas in blood, in particular the oxygen saturation of blood
US5862805A (en) Apparatus and method for measuring the variability of cardiovascular parameters
US20050038345A1 (en) Apparatus and method for non-invasive monitoring of heart performance
US5911690A (en) Use of a pulse oxymetry sensor device
US6584336B1 (en) Universal/upgrading pulse oximeter
US5782890A (en) Method for heart transplant monitoring and analog telemetry calibration
US5321492A (en) Dual function readhead for a reflectance instrument
US20080221461A1 (en) Vital sign monitor for cufflessly measuring blood pressure without using an external calibration
US8700112B2 (en) Secondary-emitter sensor position indicator
US5555882A (en) Method and apparatus for reducing ambient noise effects in electronic monitoring instruments
US5339827A (en) Acupuncture system and method
US3986496A (en) Apparatus for sensing and transmitting a pacemaker's stimulating pulse
US5257627A (en) Portable non-invasive testing apparatus
US5348004A (en) Electronic processor for pulse oximeter
US4365290A (en) Computer system with power control circuit
US4748562A (en) Portable circuit and method for performing a time study and analysis of bodily characteristics

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): CA JP

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH DE DK ES FR GB GR IE IT LU MC NL PT SE

121 Ep: the epo has been informed by wipo that ep was designated in this application
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
122 Ep: pct application non-entry in european phase
NENP Non-entry into the national phase in:

Ref country code: CA