Non-invasive ultrasonic pulse doppler cardiac output monitor

• US 5,052,395 A
• Filed: 05/16/1990
• Issued: 10/01/1991
• Est. Priority Date: 11/16/1987
• Status: Expired due to Fees

• Current Assignee: Waters Instruments, Inc. (Woodstream Corporation)
• Original Assignee: Waters Instruments, Inc. (Woodstream Corporation)
• Inventors: Davis, Donald L., Ferraro, Richard F., Sedivy, George F., Burton, Thomas A., Moehring, Mark A., Klepper, John R.
• Primary Examiner(s): Jaworski, Francis
• Assistant Examiner(s): Manuel, George
• Application Number:
  US07/499,292
• Time in Patent Office: 503 Days
• Field of Search: 128/660.1, 128/661.1, 128/61.08, 128/661.09, 128/662.01, 128/661.01
• US Class Current: 600/455
• CPC Class Codes: A61B 8/065 to determine blood output from the heart in general A61B5/029
• CPC Class Codes: A61B 8/461 Displaying means of special interest
• CPC Class Codes: A61B 8/467 characterised by special input means

Abstract

A cardiac output measurement device for the real-time, non-invasive measurement of cardiac output that can be effectively operated by relatively unskilled personnel on a routine monitoring basis in a wide variety of office and hospital conditions. To accomplish this task, the system utilizes a pulsed Doppler ultrasound transducer directed through the suprasternal notch of a patient axially towards the blood flow in the ascending aorta. The device automatically searches the ascending aorta at various predetermined depths to find the depth at which the greatest quality blood velocity reading is detected. An examination is performed at that chosen depth and the device automatically calculates a patient'"'"'s cardiac output from the Doppler measured velocity combined with an aortic diameter estimation made from the patient'"'"'s height, weight, and age. The device automatically calculates cardiac velocity, cardiac index, stroke distance, heart rate, and stroke volume.

Claims

1. A cardiac output measuring device for detecting blood flow in a patient'"'"'s ascending aorta, comprising,

an ultrasonic transducer probe adapted to be received in the patient'"'"'s suprasternal notch, including means for generating ultrasonic energy pulses directable toward the ascending aorta when the probe is positioned in the suprasternal notch, and receiving reflected frequency shifted energy pulses having a frequency shift related to the velocity of blood flow through the ascending aorta of a patient,

electronic means for separating the reflected frequency shifted signals from non-frequency shifted signals to form Doppler shift signals, and demodulating the Doppler shift signals into the normal audio frequency range,

computing means responsive to said Doppler-shift signals for computing the cardiac output of the patient,

means for selecting the depth within the aorta of the patient at which the velocity of blood flow is measured, said means comprising means for varying the time interval between the time at which a signal is transmitted and the time the reflected signal received by the ultrasonic transducer probe is selected for processing, the time intervals corresponding to varying distances from the probe and therefore also to linear positions within the ascending aorta,

pattern recognition means for testing a time series of received Doppler shift signals against predetermined signal quality characteristics and for accepting such signals as meet such characteristics, and

means responsive to said accepted signals for choosing the linear position within the ascending aorta corresponding to the greatest frequency shift.

2. The cardiac output measuring device of claim 1 including data entry means for entering height, weight and age data for the patient,

means for estimating aortic diameter from such height, weight and age data,

means for calculating heart rate using said accepted signals, and

said computing means responsive to said estimated aortic diameter and the accepted signals for computing cardiac output.

3. The cardiac output measuring device of claim 2 wherein the manual data entry means comprises an electronic keypad.

4. The cardiac output measuring device of claim 1 including manual data entry means for detecting aortic diameter data obtained from external measurements, and means responsive to the aortic diameter obtained from external measurements and the accepted frequency shifted signals for computing cardiac output.

5. The cardiac output measuring device of claim 1 wherein the means for generating and receiving ultrasonic energy pulses comprises at least one piezoelectric crystal.

6. The cardiac output measuring device of claim 1 including means for generating an audio signal having a frequency corresponding to the blood flow velocity calculated in the ascending aorta.

7. The cardiac output measuring device of claim 1 wherein the pattern recognition means for testing a time series of received Doppler-shift signals includes means for calculating Karhunen-Loeve expansion coefficients in real time and means for forming a primary feature set of signal characteristics using the Karhunen-Loeve expansion coefficients.

8. The cardiac output measuring device of claim 1 wherein the pattern recognition means for testing the time series of received Doppler-shift signals comprises:

means for calculating in real time a set of normalized Karhunen-Loeve expansion coefficients of the time series of received Doppler-shift signals;

means for calculating in real time a stroke distance and a peak velocity of the time series of received Doppler-shift signals;

means in real time for determining from the normalized Karhunen-Loeve expansion coefficients whether the time series of received Doppler-shift signals is suitable for calculation of cardiac output;

means for collecting a plurality of the suitable time series from which to calculate cardiac output;

means for computing in real time a standard deviation and means of the suitable stroke distances, and a ratio of the standard deviation to the mean;

means for computing in real time a standard deviation and mean of the suitable peak velocities, and a ratio of the standard deviation to the mean; and

means in real time for determining from the ratio of the standard deviation of the suitable peak velocities to the mean of the suitable peak velocities and from the ratio of the standard deviation of the suitable stroke distances to a mean of the suitable stroke distances, whether or not to conclude the collecting of the suitable time series, and whether or not to report the cardiac output measurement.

9. A cardiac output measuring device for detecting blood flow in a patient'"'"'s ascending aorta, comprising

(a) an ultrasonic transducer probe having a head adapted for insertion in a patient'"'"'s suprasternal notch,

(b) means for generating ultrasonic energy pulses and receiving a time series of reflected frequency shifted energy pulses having a frequency shift proportional to the velocity of blood flowing through the ascending aorta,

(c) means for receiving, and amplifying the reflected frequency shifted pulses,

(d) means for determining from the magnitude and the direction of the shift frequency, the corresponding velocity of blood flow,

(e) pattern recognition means for selection of reflected, Doppler frequency shift signals for use in calculating cardiac output, said pattern recognition means comprising means for selecting and accumulating such Doppler shift signals as qualify under all of the following criteria;

the first moment of and time derivative of such Doppler shift signals indicates blood flow toward the transducer probe in the forward direction, and for a specified period of time preceding the reception of such Doppler shift signals, signals corresponding to blood flow in the direction toward the transducer probe were not detected;

(f) means for testing accumulated time series of Doppler shift signals against predetermined signal quality characteristics and for accepting and averaging such signals as meet such characteristics,

(g) calculating means for calculating heart rate from said accumulated time series of signals,

(h) means for estimating aortic diameter from height, weight and age data,

(i) computing means responsive to said averaged signal, said heart rate and aortic diameter for computing the cardiac output of the patient, and

(j) validity testing means responsive to said heart rate and said averaged signal for determining the validity of the cardiac output measurement.

10. A cardiac output measuring device for detecting blood flow in a patient'"'"'s ascending aorta comprising

(a) an ultrasonic transducer probe receivable in a patient'"'"'s suprasternal notch, the probe having means for generating ultrasonic energy pulses directable toward blood flowing in the ascending aorta and receiving reflected frequency shifted energy pulses representative of the velocity of blood flow through the ascending aorta of a patient, said transducer having a handle adapted to be manually held by an operator, and

(b) visual display means responsive to said received frequency shifted energy pulses for displaying a visual signal representing the velocity of blood flow in the ascending aorta, the visual display signal corresponding to blood velocity of each heartbeat being overlaid upon the signal for the preceding heartbeat on the visual display means, and

(c) peak indicator means responsive to each of said blood velocity signals for producing a visual flag on said visual display means representing the peak value of each such signal, the peak flag for each signal remaining on the visual display means at least until the next heartbeat peak so that it may be compared with the peak flag position for the next succeeding signal.

11. A method for deriving an efficient set of features with which to perform real time statistical pattern recognition on time series of received Doppler-shift signals, comprising the steps of:

(a) collecting a large group of time series Doppler-shift signals;

(b) determining a set of Karhunen-Loeve expansion eigenvectors from the large number of time series of received Doppler-shift signals;

(c) collecting a second large group of time series Doppler-shift signals and separating said second group into preferred signals and unpreferred signals;

(d) calculating Karhunen-Loeve coefficients using the Karhunen-Loeve expansion eigenvectors;

(e) normalizing the Karhunen-Loeve coefficients;

(f) determining which karhunen-Loeve coefficients are statistically relevant for use in separating said preferred signals from said unpreferred signals; and

(g) determining a discriminant relation of said statistically relevant Karhunen-Loeve coefficients for use in separating said preferred signals from said unpreferred signals.