# US 11,304,662 B2 · Lung-sound signal processing method, processing device, and readable storage medium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,304,662 B2 - Document ID: US11304662B2 - Status: Granted - Kind: B2 - Application number: 15/941,117 - Filed: 2018-03-30 - Published: 2019-02-28 - Granted: 2022-04-19 - Earliest claimed: 2017-08-23 - Assignees: Boe Technology Group Co., Ltd. - Applicants: Boe Technology Group Co., Ltd. - Inventors: Xue Chen; Zhenzhong Zhang; Wei Deng - Web page: https://patentgrep.com/patent/US11304662B2 ## CPC - A61B 5/7264: {Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems} - A61B 5/08: Measuring devices for evaluating the respiratory organs - A61B 7/003: {Detecting lung or respiration noise} - G10L 25/18: the extracted parameters being spectral information of each sub-band - G10L 25/21: the extracted parameters being power information - G10L 25/66: for extracting parameters related to health condition (detecting or measuring for diagnostic purposes A61B5/00) - A61B 5/0816: {Measuring devices for examining respiratory frequency (measuring frequency of electric signals G01R23/00)} - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING ## Abstract A lung-sound signal processing method, device and computer-readable storage medium. The lung-sound processing method includes: obtaining a lung sound signal; decomposing the lung sound signal, to obtain a crackle-sound component and a respiration-sound component; obtaining an inspiratory-phase period based on the respiration-sound component; extracting a crackle-sound signal located in the inspiratory-phase period from the crackle-sound component; and obtaining at least one feature parameter according to the crackle-sound signal. ## Claims ### Claim 1 A computer-implementing lung-sound signal processing method, comprising: obtaining a lung sound signal from a lung sound signal detector; decomposing the lung sound signal by using an empirical mode decomposition algorithm, to obtain a crackle-sound component and a respiration-sound component; obtaining an inspiratory-phase period based on the respiration-sound component; extracting a crackle-sound signal located in the inspiratory-phase period from the crackle-sound component; and obtaining at least one feature parameter according to the crackle-sound signal, wherein the obtaining at least one feature parameter according to the crackle-sound signal includes: obtaining at least one time-domain feature parameter, the at least one time-domain feature parameter includes an ending time ratio (ETR) of an ending time of the crackle-sound signal to the inspiratory-phase period, wherein the ETR is calculated as: ETR = T ending T inspiration wherein Tending denotes a time duration from a starting time of a stationary respiration cycle to a time when a last crackle sound appears within the inspiratory-phase period, Tinspiration denotes a time duration of the inspiratory-phase period. ### Claim 2 (depends on 1) The lung-sound signal processing method according to claim 1, wherein the obtaining an inspiratory-phase period based on the respiration-sound component, includes: obtaining the stationary respiration cycle based on the respiration-sound component; and obtaining the inspiratory-phase period from the stationary respiration cycle; wherein the stationary respiration cycle includes an expiratory-phase period and the inspiratory-phase period. ### Claim 3 (depends on 1) The lung-sound signal processing method according to claim 1, wherein the decomposing the lung sound signal by using an empirical mode decomposition algorithm, to obtain the crackle-sound component and the respiration-sound component, includes: decomposing the lung sound signal by using an empirical mode decomposition algorithm, so that the lung sound signal is decomposed into a cumulative sum of a plurality of intrinsic mode functions; calculating a frequency spectrum of each intrinsic mode function from the plurality of intrinsic mode functions; selecting a threshold frequency; summing up intrinsic mode functions whose frequency spectra are greater than the selected threshold frequency, to obtain the crackle-sound component; and summing up intrinsic mode functions whose frequency spectra are less than or equal to the selected threshold frequency, to obtain the respiration-sound component. ### Claim 4 (depends on 3) The lung-sound signal processing method according to claim 3, wherein the selected threshold frequency is located between 30 hertz and 50 hertz. ### Claim 5 (depends on 1) The lung-sound signal processing method according to claim 1, wherein the at least one time-domain feature parameter further includes at least one of: a quantity of crackle sounds at an early stage of the inspiratory-phase period, a quantity of crackle sounds at a later stage of the inspiratory-phase period, a quantity of crackle sounds within the inspiratory-phase period, and a ratio of a time duration of the crackle-sound signal to a time length of the inspiratory-phase period. ### Claim 6 (depends on 1) The lung-sound signal processing method according to claim 1, wherein the obtaining at least one feature parameter according to the crackle-sound signal, further includes: obtaining at least one frequency-domain feature parameter. ### Claim 7 (depends on 6) The lung-sound signal processing method according to claim 6, wherein the at least one frequency-domain feature parameter includes at least one of: a total energy of the crackle-sound signal, and ratios of energies of crackle sounds of respective sub-bands to the total energy. ### Claim 8 (depends on 7) The lung-sound signal processing method according to claim 7, further comprising: obtaining a frequency spectrum function of the crackle-sound signal; obtaining a total bandwidth corresponding to the crackle-sound signal based on the frequency spectrum function; obtaining a plurality of sub-bands based on the total bandwidth; and calculating an energy of crackle sounds of each sub-band in the plurality of sub-bands. ### Claim 9 A computer readable storage medium, configured to store computer instructions, the computer instructions, when executed by a processor, causing the processor to implement one or more operations including: obtaining a lung sound signal from a lung sound signal detector; decomposing the lung sound signal by using an empirical mode decomposition algorithm, to obtain a crackle-sound component and a respiration-sound component; obtaining an inspiratory-phase period based on the respiration-sound component; extracting a crackle-sound signal located in the inspiratory-phase period from the crackle-sound component; and obtaining at least one feature parameter according to the crackle-sound signal, wherein the obtaining at least one feature parameter according to the crackle-sound signal, includes: obtaining at least one time-domain feature parameter, the at least one time-domain feature parameter includes an ending time ratio (ETR) of an ending time of the crackle-sound signal to the inspiratory-phase period, wherein the ETR is calculated as: ETR = T ending T inspiration wherein Tending denotes a time duration from a starting time of a stationary respiration cycle to a time when a last crackle sound appears within the inspiratory-phase period, Tinspiration denotes a time duration of the inspiratory-phase period. ### Claim 10 (depends on 9) The computer readable storage medium according to claim 9, wherein the computer instructions, when executed by the processor, cause the processor to further implement an operation including: classifying the crackle-sound signal based on the at least one feature parameter. ### Claim 11 A lung-sound signal processing device, comprising one or more processors, the one or more processors being configured to run computer instructions so as to execute one or more operations including: obtaining a lung sound signal from a lung sound signal detector; decomposing the lung sound signal by using an empirical mode decomposition algorithm, to obtain a crackle-sound component and a respiration-sound component; obtaining an inspiratory-phase period based on the respiration-sound component; extracting a crackle-sound signal located in the inspiratory-phase period from the crackle-sound component; and obtaining at least one feature parameter according to the crackle-sound signal, wherein the obtaining at least one feature parameter according to the crackle-sound signal, includes: obtaining at least one time-domain feature parameter, the at least one time-domain feature parameter includes an ending time ratio (ETR) of an ending time of the crackle-sound signal to the inspiratory-phase period, wherein the ETR is calculated as: E ⁢ T ⁢ R = T ending T inspiration wherein Tending denotes a time duration from a starting time of a stationary respiration cycle to a time when a last crackle sound appears within the inspiratory-phase period, Tinspiration denotes a time duration of the inspiratory-phase period. ### Claim 12 (depends on 11) The lung-sound signal processing device according to claim 11, further comprising a classifier, the classifier being configured to classify the crackle-sound signal based on the at least one feature parameter. ### Claim 13 (depends on 12) The lung-sound signal processing device according to claim 12, wherein the classifier is configured to classify the crackle-sound signal based on the at least one feature parameter by using a tree-based method. ### Claim 14 (depends on 11) The lung-sound signal processing device according to claim 11, wherein the at least one feature parameter further includes a frequency-domain feature parameter. ### Claim 15 (depends on 11) The lung-sound signal processing device according to claim 11, wherein the at least one time-domain feature parameter further includes at least one of: a quantity of crackle sounds at an early stage of the inspiratory-phase period, a quantity of crackle sounds at a later stage of the inspiratory-phase period, a quantity of crackle sounds within the inspiratory-phase period, and a ratio of a time duration of the crackle-sound signal to a time length of the inspiratory-phase period. ### Claim 16 (depends on 14) The lung-sound signal processing device according to claim 14, wherein the at least one of the frequency-domain feature parameter includes at least one of: a total energy of the crackle-sound signal, and ratios of energies of crackle sounds of respective sub-bands to the total energy. ### Claim 17 (depends on 11) The lung-sound signal processing device according to claim 11, wherein the obtaining the inspiratory-phase period based on the respiration-sound component, includes: obtaining the stationary respiration cycle based on the respiration-sound component; and obtaining the inspiratory-phase period from the stationary respiration cycle; wherein the stationary respiration cycle includes an expiratory-phase period and the inspiratory-phase period. ### Claim 18 (depends on 11) The lung-sound signal processing device according to claim 11, wherein the decomposing the lung sound signal by using an empirical mode decomposition algorithm, to obtain the crackle-sound component and the respiration-sound component, includes: decomposing the lung sound signal by using an empirical mode decomposition algorithm, so that the lung sound signal is decomposed into a cumulative sum of a plurality of intrinsic mode functions; calculating a frequency spectrum of each intrinsic mode function from the plurality of intrinsic mode functions; selecting a threshold frequency; summing up intrinsic mode functions whose frequency spectra are greater than the selected threshold frequency, to obtain the crackle-sound component; and summing up intrinsic mode functions whose frequency spectra are less than or equal to the selected threshold frequency, to obtain the respiration-sound component. ### Claim 19 (depends on 11) The lung-sound signal processing device according to claim 11, wherein the selected threshold frequency is located between 30 hertz and 50 hertz.