# US 11,304,624 B2 · Method and apparatus for performing dynamic respiratory classification and analysis for detecting wheeze particles and sources > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,304,624 B2 - Document ID: US11304624B2 - Status: Granted - Kind: B2 - Application number: 16/196,987 - Filed: 2018-11-20 - Published: 2019-06-27 - Granted: 2022-04-19 - Earliest claimed: 2012-06-18 - Assignees: AireHealth Inc. - Applicants: BreathResearch - Inventors: Charalampos-Christos Stamatopoulos; Nirinjan Bikko Yee - Web page: https://patentgrep.com/patent/US11304624B2 ## CPC - A61B 5/0823: {Detecting or evaluating cough events} - A61B 5/0205: Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition - A61B 5/024: Measuring pulse rate or heart rate - A61B 5/0816: {Measuring devices for examining respiratory frequency (measuring frequency of electric signals G01R23/00)} - A61B 5/0826: {Detecting or evaluating apnoea events (sleep apnoea A61B5/4818)} - A61B 5/6823: {Trunk, e.g., chest, back, abdomen, hip} - A61B 5/725: {using specific filters therefor, e.g. Kalman or adaptive filters (specific diagnostics methods using using bioelectric or biomagnetic signals A61B5/316)} - A61B 5/7278: {Artificial waveform generation or derivation, e.g. synthesizing signals from measured signals} - A61B 5/7465: {Arrangements for interactive communication between patient and care services, e.g. by using a telephone network (telemetry of measured physiological signal A61B5/0002)} - A61B 7/003: {Detecting lung or respiration noise} - A61B 7/04: Electric stethoscopes - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/08: Learning methods - G06N 3/09: Supervised learning - G10L 25/66: for extracting parameters related to health condition (detecting or measuring for diagnostic purposes A61B5/00) - G16H 20/30: relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising - G16H 40/60: for the operation of medical equipment or devices - A61B 5/6819: {Nose} - A61B 5/6831: {Straps, bands or harnesses} - A61B 5/742: {using visual displays (displays for heart-related electrical signals, e.g. ECG, A61B5/339)} - G06N 3/045: Combinations of networks - G10L 25/06: the extracted parameters being correlation coefficients - G10L 25/18: the extracted parameters being spectral information of each sub-band - G10L 25/30: using neural networks - G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA ## Abstract A method for analyzing an audio respiratory signal comprises capturing the audio respiratory signal from a subject using a microphone and partitioning the audio respiratory signal into a plurality of overlapping frames. The method further comprises calculating a fourier transform for each frame and determining a magnitude spectrum using the fourier transform of the plurality of overlapping frames. Additionally, the method comprises extracting a spectrogram using the magnitude spectrum and analyzing the spectrogram to determine characteristics pertaining to wheeze sounds in the audio respiratory signal. ## Claims ### Claim 1 A method for analyzing an audio respiratory signal, the method comprising: capturing the audio respiratory signal from a subject using a microphone; partitioning the audio respiratory signal into a plurality of overlapping frames and a plurality of non-overlapping frames; calculating a fourier transform for each frame; determining a magnitude spectrum using the fourier transform of the plurality of overlapping frames; extracting a spectrogram using the magnitude spectrum; analyzing the spectrogram to determine characteristics pertaining to wheeze sounds in the audio respiratory signal; creating an artificial crackle comprising a filtered impulse response frame created by filtering a delta function; calculating a cross correlation function between each of the non-overlapping frames and the artificial crackle; determining if the cross correlation function exceeds a predetermined threshold; and responsive to a determination that the cross correlation function exceeds the predetermined threshold, identifying a respective frame for which the cross correlation function exceeds the predetermined threshold as associated with crackling. ### Claim 2 (depends on 1) The method of claim 1, wherein the artificial crackle is created by filtering the delta function in the filtered impulse response frame with a narrow IIR bandpass filter. ### Claim 3 (depends on 1) The method of claim 1, further comprising: storing a timestamp and intensity corresponding to each frame identified as associated with crackling; and extracting a plurality of characteristics pertaining to crackling using frames identified as associated with crackling. ### Claim 4 (depends on 1) The method of claim 1, further comprising: extracting a magnified spectrogram using the spectrogram, wherein the magnified spectrogram is operable to determine a number of wheeze particles. ### Claim 5 (depends on 1) The method of claim 1, further comprising: extracting an interpolated spectrogram using the spectrogram, wherein the interpolated spectrogram is operable to provide a clearer view of wheeze particles by subtracting out residual airflow energies from the spectrogram. ### Claim 6 (depends on 1) The method of claim 1, wherein the analyzing further comprises: executing an edge-detection procedure on the spectrogram to highlight wheeze particles in the spectrogram. ### Claim 7 (depends on 6) The method of claim 6, wherein the analyzing further comprises: for each column in the spectrogram, performing the following: (a) store locations of elements with high values resulting from the edge-detection procedure into a vector; (b) compare a location of the first element in the vector with locations of remaining elements in the vector; (c) designate any of the remaining elements where a respective location is a multiple of the location of the first element as a harmonic of the first element; and (d) repeat steps a), b) and c) for each element in the vector. ### Claim 8 A non-transitory computer-readable storage medium having stored thereon, computer executable instructions that, if executed by a computer system cause the computer system to perform a method for analyzing an audio respiratory signal, the method comprising: capturing the audio respiratory signal from a subject using a microphone; partitioning the audio respiratory signal into a plurality of overlapping frames and a plurality of non-overlapping frames; calculating a fourier transform for each frame; extracting a spectrogram using the fourier transform of the plurality of overlapping frames; analyzing the spectrogram to determine characteristics pertaining to wheeze sounds in the audio respiratory signal; creating an artificial crackle comprising a filtered impulse response frame created by filtering a delta function; calculating a cross correlation function between each of the non-overlapping frames and the artificial crackle; determining if the cross correlation function exceeds a predetermined threshold; and responsive to a determination that the cross correlation function exceeds the predetermined threshold, identifying a respective frame for which the cross correlation function exceeds the predetermined threshold as associated with crackling. ### Claim 9 (depends on 8) The non-transitory computer-readable storage medium of claim 8, wherein the artificial crackle is created by filtering the delta function in the filtered impulse response frame with a narrow IIR bandpass filter. ### Claim 10 (depends on 8) The non-transitory computer-readable storage medium of claim 8, wherein the method further comprises: storing a timestamp and intensity corresponding to each frame identified as associated with crackling; and extracting a plurality of characteristics pertaining to crackling using frames identified as associated with crackling. ### Claim 11 (depends on 8) The non-transitory computer-readable storage medium of claim 8, wherein the method further comprises: extracting a magnified spectrogram using the spectrogram, wherein the magnified spectrogram is operable to determine a number of wheeze particles. ### Claim 12 (depends on 8) The non-transitory computer-readable storage medium of claim 8, wherein the method further comprises: extracting an interpolated spectrogram using the spectrogram, wherein the interpolated spectrogram is operable to provide a clearer view of wheeze particles by subtracting out residual airflow energies from the spectrogram. ### Claim 13 A system for detecting wheeze and crackle from an audio respiratory signal, the system comprising: a spirometer comprising a first microphone, wherein the first microphone is operable to capture the audio respiratory signal from a subject; a memory coupled to the spirometer and operable to store the audio respiratory signal, wherein the memory further comprises an application for detecting wheeze and crackle from a breathing session stored therein; and a processor coupled to said memory and said spirometer, the processor being configured to operate in accordance with said application to: capture the audio respiratory signal from a subject using a microphone; partition the audio respiratory signal into a plurality of overlapping frames and a plurality of non-overlapping frames; calculate a fourier transform for each frame; determine a magnitude spectrum using the fourier transform of the plurality of overlapping frames; extract a spectrogram using the magnitude spectrum; analyze the spectrogram to determine characteristics pertaining to wheeze sounds in the audio respiratory signal; create an artificial crackle comprising a filtered impulse response frame created by filtering a delta function; calculate a cross correlation function between each of the non-overlapping frames and the artificial crackle; determine if the cross correlation function exceeds a predetermined threshold; and responsive to a determination that the cross correlation function exceeds the predetermined threshold, identify a respective frame for which the cross correlation function exceeds the predetermined threshold as associated with crackling. ### Claim 14 (depends on 13) The system of claim 13, wherein the artificial crackle is created by filtering the delta function in the filtered impulse response frame with a narrow IIR bandpass filter. ### Claim 15 (depends on 13) The system of claim 13, wherein the processor is further configured to: store a timestamp and intensity corresponding to each frame identified as associated with crackling; and extract a plurality of characteristics pertaining to crackling using frames identified as associated with crackling. ### Claim 16 (depends on 13) The system of claim 13, wherein the processor is further configured to: extract a magnified spectrogram using the spectrogram, wherein the magnified spectrogram is operable to determine a number of wheeze particles. ### Claim 17 (depends on 13) The system of claim 13, wherein the processor is further configured to: extract an interpolated spectrogram using the spectrogram, wherein the interpolated spectrogram is operable to provide a clearer view of wheeze particles by subtracting out residual airflow energies from the spectrogram.