Predicting characteristics of a future respiratory event, and applications thereof
AssigneeSonavi Labs, Inc.
InventorIan Mitra McLane
Life of this patent
Estimating the term…
Abstract
Embodiments disclosed herein improve digital stethoscopes and their application and operation. A first method detects of a respiratory abnormality using a convolution. A second method counts coughs for a patient. A third method predicts a respiratory event based on a detected trend. A fourth method forecasts characteristics of a future respiratory event. In a fifth embodiment, a base station is provided for a digital stethoscope.
Claims14
A computer-implemented method for forecasting characteristics of a future respiratory event comprising:
- receiving, from a microphone of a digital stethoscope, a first noise signal capturing a first respiratory event at a first time;
- generating, by one or more computing devices, a first feature space representing the first noise signal;
- encoding, by the one or more computing devices, the first feature space into a first convolution vector using a spatial convolution procedure;
- receiving, from the microphone of the digital stethoscope, a second noise signal capturing a second respiratory event at a second time;
- generating, by the one or more computing devices, a second feature space representing the second noise signal;
- encoding, by the one or more computing devices, the second feature space into a second convolution vector using the spatial convolution procedure;
- determining, by the one or more computing devices, a predicted convolution vector based on the first and second feature spaces;
- decoding, by the one or more computing devices, the predicted convolution vector into a predicted feature space representing a sound made by the future respiratory event;
- and
- transmitting, by the one or more computing devices, the predicted feature space to a display unit for output, wherein the output indicates the likelihood of the future respiratory event.
A non-transitory computer readable medium including instructions for forecasting characteristics of a future respiratory event comprising:
- receiving, from a microphone of a digital stethoscope, a first noise signal capturing a first respiratory event at a first time;
- generating, by one or more computing devices, a first feature space representing the first noise signal;
- encoding, by the one or more computing devices, the first feature space into a first convolution vector using a spatial convolution procedure;
- receiving, from the microphone of the digital stethoscope, a second noise signal capturing a second respiratory event at a second time;
- generating, by the one or more computing devices, a second feature space representing the second noise signal;
- encoding, by the one or more computing devices, the second feature space into a second convolution vector using the spatial convolution procedure;
- determining, by the one or more computing devices, a predicted convolution vector based on the first and second feature spaces;
- decoding, by the one or more computing devices, the predicted convolution vector into a predicted feature space representing a sound made by the future respiratory event;
- and
- transmitting, by the one or more computing devices, the predicted feature space to a display unit for output, wherein the output indicates the likelihood of the future respiratory event.
Dependencies are inferred from the claim text.