System and method for noninvasive detection of arterial stenosis

• US 5,638,823 A
• Filed: 08/28/1995
• Issued: 06/17/1997
• Est. Priority Date: 08/28/1995
• Status: Expired due to Fees

• Current Assignee: Rutgers University
• Original Assignee: Rutgers University
• Inventors: Akay, Metin, Welkowitz, Walter, Akay, Yasemin M., Kostis, John
• Primary Examiner(s): Nasser, Robert L.
• Application Number:
  US08/520,236
• Time in Patent Office: 659 Days
• Field of Search: 128/668, 128/670, 128/691, 128/687, 128/710, 128/713, 128/715, 128/690
• US Class Current: 600/528
• CPC Class Codes: A61B 5/02007 Evaluating blood vessel condition, e.g. elasticity, compliance
• CPC Class Codes: A61B 5/726 using Wavelet transforms
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: A61B 5/7275 Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: A61B 7/00 Instruments for auscultation
• CPC Class Codes: A61B 7/04 Electric stethoscopes microphones, acoustic transducers therefor H04R
• CPC Class Codes: G16H 40/63 for local operation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment

Abstract

A system and method for noninvasively detecting coronary artery disease. The system and method utilize a vasodilator drug to increase the signal-to-noise ratio of an acoustic signal that represents diastolic heart sounds of a patient. A wavelet transform is performed on the acoustic signal to provide parameters for a feature vector. Scaled clinical examination parameters such as a patient'"'"'s sex, age, body weight, smoking condition, blood pressure, and family history are also included in the feature vector. The feature vector is used as an input pattern to neural networks. The output of the neural networks represent a diagnosis of coronary stenosis in a patient.

Claims

1. A method for generating a risk indicator representing a probability of a patient having an arterial stenosis, comprising the steps of:

receiving a first signal representing one or more sound event caused by turbulence of blood flowing in an artery;

performing a wavelet transform on said first signal to provide information on both a frequency localization and a temporal localization of said one or more sound events represented by said first signal;

generating a feature vector having a plurality of parameters, said parameters include parameters resulting from the wavelet transform and a clinical examination; and

generating the risk indicator by manipulating said feature vector using a neural network.

2. The method of claim 1, further comprising the step of:

improving a signal-to-noise ratio of said first signal before said first signal is received by increasing a rate of blood flow in said artery of the patient.

3. The method of claim 2, wherein said step of improving said signal-to-noise ratio comprises an application of a vasodilator drug.

4. The method of claim 2, wherein said first signal represents the sound of turbulence caused by blood flowing through an artery during a diastole.

5. The method of claim 4, wherein said artery is a coronary artery.

6. The method of claim 2, wherein said artery is a carotid artery.

7. The method of claim 1, wherein said step of generating the risk indicator includes:

receiving said feature vector;

using said feature vector as an input to a neural network; and

determining the risk indicator based upon an output of said neural network.

8. The method of claim 7, wherein said neural network is a fuzzy minimum-maximum neural network.

9. A computer-based system for determining a risk indicator representing a probability of a patient having an arterial stenosis, comprising:

a processor;

a storage medium including;

a wavelet transform determinator, coupled to said processor, said wavelet transform determinator having an input coupled to receive a first signal representing acoustic measurements from blood flowing through an artery, and having an output, for performing a wavelet transform upon said first signal to generate a second signal at said wavelet transform determinator output representing acoustic events in said first signal;

a feature vector generator, having an input coupled to said wavelet transform determinator output, for generating a feature vector based upon said wavelet transform determinator output; and

a neural network determinator, having an input coupled to said feature vector generator, for receiving said feature vector and processing said feature vector with a neural network.

10. The system of claim 9, wherein said feature vector generator utilizes scaled clinical examination data to generate said feature vector.

11. The system of claim 10, further comprising a clinical data scaler, coupled to said processor, for modifying clinical examination data to generate scaled clinical examination data to be input into said neural network determinator.

12. The system of claim 11, wherein said clinical examination data includes a sex of the patient and a blood pressure of the patient.

13. The system of claim 9, wherein said wavelet transform determinator transforms said first signal to provide both frequency and temporal localization of acoustic events represented by said first signal.