Detection of coronary artery disease using an electronic stethoscope
The disclosure describes an electronic stethoscope system that automatically detects coronary artery disease in patients. The system uses an electronic stethoscope to record acoustic data from the fourth left intercostal space of a patient. A processing technique is then applied in order to filter the data and produce Fast Fourier Transform (FFT) data of magnitude versus frequency. If a bell curve is identified in the data between a predefined frequency range (e.g., 50 and 80 Hz) with a peak magnitude of greater than a predefined threshold (e.g., 2.5 units), the system automatically provides an output indicating that the patient is likely to have 50 to 99 percent stenosis of the coronary artery. If no bell curve is present, the patient may have artery stenosis of less than 50 percent. An interface module may be used to transfer diagnosis information to the stethoscope and data to a general purpose computer. This inexpensive and quick system may improve the chances for early detection and patient survival of coronary artery disease.
1. A method for detecting coronary artery disease using an electronic stethoscope, the method comprising:
recording acoustic data with an electronic stethoscope having an acoustic sensor placed over a fourth left intercostal space of a patient's chest;
applying one or more filters to the acoustic data and calculating a Fast Fourier Transform (FFT) of the data to produce FFT data;
analyzing the FFT data to detect a bell curve within a predefined frequency range of 50 and 80 Hz indicative of coronary artery disease; and
generating an output on a display when the bell curve is detected to indicate that the patient is likely to have coronary artery disease.
2. The method of claim 1 , wherein analyzing the FFT data includes analyzing the FFT data with the electronic stethoscope to identify the bell curve.
3. The method of claim 1 , wherein generating an output comprises updating the display of the electronic stethoscope.
4. The method of claim 1 , further comprising communicating the acoustic data to an external computer via an interface module, and
wherein the steps of applying one or more filters and analyzing the FFT data are performed within the external computer.
5. The method of claim 1 , wherein the acoustic sensor is a microphone.
6. The method of claim 1 , wherein the acoustic sensor is a piezoelectric crystal.
7. The method of claim 1 , wherein the acoustic data includes sounds generated during systole and diastole.
8. The method of claim 1 , further comprising sending the acoustic data recorded with the electronic stethoscope to a general purpose computer for analyzing and outputting the indicator.
9. The method of claim 1 , wherein one filter removes frequencies not related to stenosis from the acoustic data.
10. The method of claim 1 , wherein one filter is a low pass filter that removes frequencies higher than 100 Hz.
11. The method of claim 1 , further comprising displaying a graph of the processed data for manual diagnosis including plotting harmonic magnitudes on the y-axis and plotting harmonic frequencies on the x-axis.
12. The method of claim 11 , wherein displaying the graph comprises:
displaying the graph in a first color, and
displaying the detected bell curve in a second color to aid recognition by a clinician.
13. The method of claim 1 , further comprising identifying the onset of the bell curve on or after 50 Hz.
14. The method of claim 1 , further comprising identifying a downslope in the curve that ends at or before 80 Hz.
15. The method of claim 1 , further comprising identifying a FFT peak in the bell curve greater than 2.5 units.
16. The method of claim 1 , further comprising:
computing a total energy associated with the bell curve;
determining a percentage indicator of an extent of the artery stenosis based on the total energy; and
outputting the percentage indicator.
17. The method of claim 1 , further comprising:
computing a total area under the bell curve;
determining a percentage indicator of an extent of the artery stenosis based on the total area; and
outputting the percentage indicator.
18. The method of claim 1 , wherein the coronary artery disease is detected in the left anterior descending coronary artery.
19. An electronic stethoscope comprising:
a display;
an acoustic sensor;
one or more memories that stores acoustic data from the acoustic sensor placed over a fourth left intercostal space of a patient's chest and instructions for processing the acoustic data; and
a processor that applies one or more filters to the acoustic data and calculates a Fast Fourier Transform (FFT) of the data to produce FFT data,
wherein the processor analyzes the FFT data to detect a bell curve having an onset on or after 50 Hz and a downslope at or before 80 Hz, and generates an output on the display when the bell curve is detected to indicate that the patient is likely to have coronary artery disease.
20. The electronic stethoscope of claim 19 , wherein the processor analyzes the EFT data to identify a peak in the bell curve greater than 2.5 units.
21. The electronic stethoscope of claim 19 , wherein the processor analyzes the FFT data to compute a total energy associated with the bell curve, determine a percentage indicator of an extent of the artery stenosis based on the total energy, and outputs the percentage indicator.
22. The electronic stethoscope of claim 19 , wherein the processor analyzes the FFT data to computing a total area under the bell curve, determine a percentage indicator of an extent of the artery stenosis based on the total area; and output the percentage indicator.
23. A computer-readable medium comprising instructions to cause an electronic stethoscope having a processor to:
record acoustic data with an electronic stethoscope having an acoustic sensor placed over a fourth left intercostal space of a patient's chest;
apply one or more filters to the acoustic data and calculates a Fast Fourier Transform (FFT) of the data to produce FFT data;
analyze the FFT data to detect a bell curve between 50 and 80 Hz; and
generate an output on a display when the bell curve is detected to indicate that the patient is likely to have coronary artery disease.
24. A method for detecting coronary artery disease using an electronic stethoscope, the method comprising:
recording acoustic data with an electronic stethoscope having an acoustic sensor placed over a fourth left intercostal space of a patient's chest;
applying one or more filters to the acoustic data and calculating a Fast Fourier Transform (FET) of the data to produce FFT data;
displaying a graph of the FFT data for manual diagnosis including plotting harmonic magnitudes on the y-axis and plotting harmonic frequencies on the x-axis;
analyzing the FFT data to detect a bell curve within a predefined frequency range indicative of coronary artery disease; and
generating an output on a display when the bell curve is detected to indicate that the patient is likely to have coronary artery disease.