SYSTEM AND METHOD FOR DIGITAL AUSCULTATION
The system can include: a sphygmomanometer and a stethoscope. The method can include: collecting data; processing the data; and determining a cardiovascular parameter value. In variants, the system and/or method can function to automatically determine a cardiovascular parameter value for a user from auscultation data (e.g., without manually listening to the auscultation data). Additionally or alternatively, the system and/or method can function to validate a cardiovascular parameter device (e.g., a user device).
1 . A system, comprising:
a sphygmomanometer configured to measure pressure data at an arm region of a user;
a stethoscope configured to measure audio data at the arm region of the user;
a data recorder configured to simultaneously record the pressure data from the sphygmomanometer and the audio data from the stethoscope; and
a computing system comprising non-transitory computer-readable media, storing computer-readable instructions that, when executed by the computing system, cause the computing system to:
determine a spectrogram from the audio data;
convert the spectrogram into an image;
identify potential heartbeats by detecting heartbeat patterns in the image;
segment the potential heartbeats into a set of segments;
filter the potential heartbeats to identify true heartbeats, wherein the potential heartbeats are filtered based on at least one of: a number of potential heartbeats in each segment of the set of segments or a potential pulse rate of potential heartbeats in each segment of the set of segments; and
determine a cardiovascular parameter value based on the true heartbeats and the pressure data.
2 . The system of claim 1 , wherein segmenting the potential heartbeats comprises: identifying time gaps between adjacent potential heartbeats that are greater than a threshold time gap, and segmenting the potential heartbeats such that adjacent segments in the set of segments are separated by the identified time gaps.
3 . The system of claim 2 , wherein the threshold time gap is determined based on an overall pulse rate, the overall pulse rate determined based on the pressure data.
4 . The system of claim 1 , wherein filtering the potential heartbeats to identify the true heartbeats comprises: performing a first filtering operation on the set of segments to identify a first set of filtered heartbeats, segmenting the first set of filtered heartbeats into a second set of segments, and performing a second filtering operation on the second set of segments to identify the true heartbeats.
5 . The system of claim 4 , wherein time gaps between adjacent segments of the set of segments are larger than time gaps between adjacent segments of the second set of segments.
6 . The system of claim 1 , wherein filtering the potential heartbeats comprises: determining a number of potential heartbeats in each segment of the set of segments, and removing potential heartbeats corresponding to each segment with a number of potential heartbeats that is less than a threshold number.
7 . The system of claim 1 , wherein filtering the potential heartbeats comprises: determining a potential pulse rate for each segment of the set of segments, and removing potential heartbeats corresponding to each segment with a potential pulse rate that is outside a target pulse rate range.
8 . The system of claim 1 , wherein the heartbeat patterns comprise line objects that are within a threshold angle of vertical and are greater than a threshold length.
9 . The system of claim 1 , further comprising a cardiovascular parameter device coupled to the user at a second arm region of the user, wherein the cardiovascular parameter value is used to validate the cardiovascular parameter device.
10 . The system of claim 1 , wherein the cardiovascular parameter value comprises at least one of: a systolic blood pressure value, a diastolic blood pressure value, or a mean arterial pressure value.
11 . A method, comprising:
measuring pressure data at an arm region of a user using a sphygmomanometer;
measuring audio data at the arm region of the user using a stethoscope;
processing the audio data, comprising:
determining a spectrogram from the audio data;
converting the spectrogram into an image;
identifying potential heartbeats in the image;
segmenting the potential heartbeats into a set of segments; and
filtering the potential heartbeats to identify true heartbeats, wherein the potential heartbeats are filtered based on at least one of: a number of potential heartbeats in each segment of the set of segments or a pulse rate of potential heartbeats in each segment of the set of segments; and
determining a cardiovascular parameter value based on the true heartbeats and the pressure data.
12 . The method of claim 11 , wherein the image comprises a set of windows corresponding to windows of the spectrogram, the method further comprising determined a noise-reduced image by transforming each window of the image based on an aggregate intensity of the window, wherein the potential heartbeats are identified using the noise-reduced image.
13 . The method of claim 11 , further comprising providing a therapeutic intervention based on the cardiovascular parameter value.
14 . The method of claim 11 , wherein segmenting the potential heartbeats comprises: identifying time gaps between adjacent potential heartbeats that are greater than a threshold time gap, the threshold time gap determined based on an overall pulse rate, the overall pulse rate determined based on the pressure data; and segmenting the potential heartbeats such that adjacent segments in the set of segments are separated by the identified time gaps.
15 . The method of claim 11 , wherein filtering the potential heartbeats to identify the true heartbeats comprises: performing a first filtering operation on the set of segments to identify a first set of filtered heartbeats, segmenting the first set of filtered heartbeats into a second set of segments, and performing a second filtering operation on the second set of segments to identify the true heartbeats.
16 . The method of claim 15 , wherein time gaps between adjacent segments of the set of segments are larger than time gaps between adjacent segments of the second set of segments.
17 . The method of claim 11 , further comprising: performing a set of pressure data checks on the pressure data and performing a set of audio data checks on the audio data, wherein the audio data is processed in response to the pressure data passing the pressure data checks and the audio data passing the audio data checks.
18 . The method of claim 17 , wherein the pressure data checks comprise: a sphygmomanometer deflation rate check, a sphygmomanometer deflation timing check, a pressure data linearity check, and a mean arterial pressure check.
19 . The method of claim 17 , wherein the audio data and the pressure data are measured during an inflation period and during a deflation period, wherein the audio data checks comprise: a first audio power check at a start of the deflation period, a second audio power check at an end of the deflation period, and a noise check.
20 . The method of claim 11 , wherein the cardiovascular parameter value comprises at least one of: a systolic blood pressure value, a diastolic blood pressure value, or a mean arterial pressure value.