# US 10,765,399 B2 · Programmable electronic stethoscope devices, algorithms, systems, and methods > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,765,399 B2 - Document ID: US10765399B2 - Status: Granted - Kind: B2 - Application number: 15/772,454 - Filed: 2016-10-31 - Published: 2018-11-08 - Granted: 2020-09-08 - Earliest claimed: 2015-10-30 - Assignees: The Johns Hopkins University - Applicants: The Johns Hopkins University - Inventors: Dimitra Emmanouilidou; James E. West; Mounya Elhilali; Ian Mclane - Web page: https://patentgrep.com/patent/US10765399B2 ## CPC - A61B 7/04: Electric stethoscopes - A61B 5/08: Measuring devices for evaluating the respiratory organs - A61B 5/7203: {for noise prevention, reduction or removal} - A61B 5/7264: {Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems} - A61B 7/003: {Detecting lung or respiration noise} - A61B 7/026: {comprising more than one sound collector} - G06N 20/00: Machine learning - G06N 99/00: Subject matter not provided for in other groups of this subclass - G16H 40/63: for local operation - G10K 2210/116: Medical; Dental - G10L 25/51: for comparison or discrimination - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA - G10K: SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING ## Abstract A digital electronic stethoscope includes an acoustic sensor assembly that includes a body sensor portion and an ambient sensor portion, the body sensor portion being configured to make acoustically coupled contact with a subject while the ambient sensor portion is configured to face away from the body sensor portion so as to capture environmental noise proximate the body sensor portion; a signal processor and data storage system configured to communicate with the acoustic sensor assembly so as to receive detection signals therefrom, the detection signals including an auscultation signal comprising body target sound and a noise signal; and an output device configured to communicate with the signal processor and data storage system to provide at least one of an output signal or information derived from the output signal. The signal processor and data storage system includes a noise reduction system that removes both stationary noise and non-stationary noise from the detection signal to provide a clean auscultation signal substantially free of distortions. The signal processor and data storage system further includes an auscultation sound classification system configured to receive the clean auscultation signal and provide a classification thereof as at least one of a normal breath sound or an abnormal breath sound. ## Claims ### Claim 1 A digital electronic stethoscope, comprising: an acoustic sensor assembly comprising a body sensor portion and an ambient sensor portion, said body sensor portion being configured to make acoustically coupled contact with a subject while said ambient sensor portion is configured to face away from said body sensor portion so as to capture environmental noise proximate said body sensor portion; a signal processor and data storage system configured to communicate with said acoustic sensor assembly so as to receive detection signals therefrom, said detection signals comprising an auscultation signal comprising body target sound and a noise signal; and an output device configured to communicate with said signal processor and data storage system to provide at least one of an output signal or information derived from said output signal, wherein said signal processor and data storage system comprises a noise reduction system that removes both stationary noise and non-stationary noise from said detection signal to provide a clean auscultation signal substantially free of distortions, wherein said signal processor and said data storage system further comprises an auscultation sound classification system configured to receive said clean auscultation signal and provide a classification thereof as at least one of a normal breath sound or an abnormal breath sound, and wherein said noise reduction system suppresses ambient noise by multiband spectral subtraction of said noise signal from said detection signal to provide said clean auscultation signal substantially free of distortions. ### Claim 2 (depends on 1) The digital electronic stethoscope according to claim 1, wherein said body sensor portion of said acoustic sensor assembly comprises a microphone array of a plurality of microphones. ### Claim 3 (depends on 2) The digital electronic stethoscope according to claim 2, wherein said each of said plurality of microphones is an electret microphone. ### Claim 4 (depends on 1) The digital electronic stethoscope according to claim 1, wherein said output device is at least one of earphones, a smart phone, or a computer. ### Claim 5 (depends on 1) The digital electronic stethoscope according to claim 1, wherein said noise reduction system comprises a clipping repair system to repair clipping of said auscultation signal to provide said clean auscultation signal substantially free of distortions. ### Claim 6 (depends on 1) The digital electronic stethoscope according to claim 1, wherein said noise reduction system comprises a heart sound elimination system to remove said subject's heart sounds from said detection signal to provide said clean auscultation signal substantially free of distortions. ### Claim 7 (depends on 1) The digital electronic stethoscope according to claim 1, wherein said noise reduction system comprises a friction noise removal system to remove friction noise of said acoustic sensor assembly rubbing against said subject from said detection signal to provide said clean auscultation signal substantially free of distortions. ### Claim 8 (depends on 6) The digital electronic stethoscope according to claim 6, wherein said noise reduction system comprises a friction noise removal system to remove friction noise of said acoustic sensor assembly rubbing against said subject from said detection signal to provide said clean auscultation signal substantially free of distortions. ### Claim 9 (depends on 1) The digital electronic stethoscope according to claim 1, wherein said noise reduction system comprises a subject's crying removal system to remove said crying noise of said subject from said detection signal to provide said clean auscultation signal substantially free of distortions. ### Claim 10 (depends on 7) The digital electronic stethoscope according to claim 7, wherein said noise reduction system comprises a subject's crying removal system to remove said crying noise of said subject from said detection signal to provide said clean auscultation signal substantially free of distortions. ### Claim 11 (depends on 8) The digital electronic stethoscope according to claim 8, wherein said noise reduction system comprises a subject's crying removal system to remove said crying noise of said subject from said detection signal to provide said clean auscultation signal substantially free of distortions. ### Claim 12 (depends on 1) The digital electronic stethoscope according to claim 1, wherein said auscultation sound classification system is a machine learning system that learns from training data. ### Claim 13 (depends on 11) The digital electronic stethoscope according to claim 11, wherein said auscultation sound classification system uses a binary support vector machine (SVM) algorithm and radial-basis kernels (RBFs). ### Claim 14 (depends on 1) The digital electronic stethoscope according to claim 1, wherein said multiband spectral subtraction includes: processing individual frequency bands based on spectral characteristics of a target body sound and said noise signal, adjusting processing of a localized time window of said detection signal based on local Signal To Noise Ratio (SNR) information to provide a processed signal, and smoothing said processed signal along adjacent time frames and frequency bands to reduce reconstruction distortions in said processed signal. ### Claim 15 A method of processing signals detected by a digital electronic stethoscope, comprising: obtaining an auscultation signal from said electronic stethoscope, said auscultation signal comprising a target body sound; obtaining a noise signal comprising noise from an environment of said body; obtaining a processed signal by reducing unwanted noise in said auscultation signal based on at least one of said auscultation signal and said noise signal; performing acoustic analysis of said processed signal; and performing statistical analysis of said processed signal, wherein said obtaining said processed signal includes suppressing ambient noise by multiband spectral subtraction of said noise signal from said auscultation signal. ### Claim 16 (depends on 15) The method of processing signals according to claim 15, wherein said obtaining said processed signal includes detecting and repairing clipping distortion in said auscultation signal. ### Claim 17 (depends on 15) The method of processing signals according to claim 15, wherein said multiband spectral subtraction includes: processing individual frequency bands based on spectral characteristics of said target body sound and said noise signal, adjusting the processing of a localized time window of said auscultation signal based on local Signal To Noise Ratio (SNR) information, and smoothing said processed signal along adjacent time frames and frequency bands to reduce reconstruction distortions in said processed signal. ### Claim 18 (depends on 15) The method of processing signals according to claim 15, wherein said obtaining said processed signal includes reducing mechanical noise of said electronic stethoscope. ### Claim 19 (depends on 15) The method of processing signals according to claim 15, wherein said obtaining said processed signal includes suppressing noise from crying of a patient from whom said auscultation signal was obtained. ### Claim 20 (depends on 15) The method of processing signals according to claim 15, wherein said obtaining said processed signal includes: identifying and suppressing or extracting heart sounds in said auscultation signal, and extrapolating said processed signal across intervals where said heart sounds were suppressed or extracted using an auto-regressive/moving average (ARMA) method. ### Claim 21 (depends on 15) The method of processing signals according to claim 15, wherein said acoustic analysis includes: obtaining an auditory spectrogram by enhancing said processed signal using a plurality of filters, and capturing signal modulations along both time and frequency axes of said auditory spectrogram. ### Claim 22 (depends on 15) The method of processing signals according to claim 15, wherein said statistical analysis includes classification of sounds obtained from said auscultation signal using binary support vector machines (SVMs) and radial-basis kernels (RBFs). ### Claim 23 A computer-readable medium comprising non-transitory computer-executable code for processing signals detected by a digital electronic stethoscope, which when executed by a computer causes the computer to: obtain an auscultation signal from said electronic stethoscope, said auscultation signal comprising a target body sound; obtain a noise signal comprising noise from an environment of said body; obtain a processed signal by reducing unwanted noise in said auscultation signal based on at least one of said auscultation signal and said noise signal; performing acoustic analysis of said processed signal; and performing statistical analysis of said processed signal, wherein said obtaining said processed signal includes suppressing ambient noise by multiband spectral subtraction of said noise signal from said auscultation signal. ### Claim 24 (depends on 23) The computer-readable medium according to claim 23, wherein said obtaining said processed signal includes detecting and repairing clipping distortion in said auscultation signal. ### Claim 25 (depends on 23) The computer-readable medium according to claim 23, wherein said multiband spectral subtraction includes: processing individual frequency bands based on spectral characteristics of said target body sound and said noise signal, adjusting the processing of a localized time window of said auscultation signal based on local Signal To Noise Ratio (SNR) information, and smoothing said processed signal along adjacent time frames and frequency bands to reduce reconstruction distortions in said processed signal. ### Claim 26 (depends on 23) The computer-readable medium according to claim 23, wherein said obtaining said processed signal includes reducing mechanical noise of said electronic stethoscope. ### Claim 27 (depends on 23) The computer-readable medium according to claim 23, wherein said obtaining said processed signal includes suppressing noise from crying of a patient from whom said auscultation signal was obtained. ### Claim 28 (depends on 23) The computer-readable medium according to claim 23, wherein said obtaining said processed signal includes: identifying and suppressing or extracting heart sounds in said auscultation signal, and extrapolating said processed signal across intervals where said heart sounds were suppressed or extracted using an auto-regressive/moving average (ARMA) method. ### Claim 29 (depends on 23) The computer-readable medium according to claim 23, wherein said acoustic analysis includes: obtaining an auditory spectrogram by enhancing said processed signal using a plurality of filters, and capturing signal modulations along both time and frequency axes of said auditory spectrogram. ### Claim 30 (depends on 23) The computer-readable medium according to claim 23, wherein said statistical analysis includes classification of sounds obtained from said auscultation signal using binary support vector machines (SVMs) and radial-basis kernels (RBFs).