Device embedded in, or attached to, a pillow configured for in-bed monitoring of respiration
A monitoring device has microphones, an ADC; a digital radio; and a processor with firmware. The firmware includes code for digitizing audio from the microphones into time-domain audio, performing FFT to provide frequency-domain audio, running a first neural network on time domain and frequency-domain audio to extract features, executing a classifier on the features to identify candidate events, and using the digital radio to upload candidate events and features. A pressure sensor awakens the processor from a low-power state. In particular embodiments, the first neural network is an embedded Gated Recurrent Unit having weights trained to extract features of use in the classifier; and candidate events include normal inhalation and exhalation breathing sounds, crackles, wheezes, coughs, snoring, gasping, choking, and speech sounds and in some embodiments heart sounds. A method of monitoring breathing during sleep includes attaching the device to, or embedding the device within, a pillow.
1. A device configured for monitoring physiological sounds comprising:
at least one microphone coupled to an analog-to-digital converter (ADC);
a digital radio;
a processor configured with firmware in a memory, the firmware comprising machine readable code configured to:
use the ADC to digitize audio signals from the at least one microphone into digitized time-domain audio signals,
perform a fast Fourier transform on the digitized time-domain audio signals to provide frequency-domain audio signals,
execute a first neural network on at least the frequency-domain audio signals to extract features from the frequency-domain audio signals and at least one pressure sensor,
execute second neural network configured to classify events on the features to identify candidate events, and
use the digital radio to upload the candidate events and features;
where the at least one pressure sensor is coupled through a wake-up circuit configured to awaken the processor.
2. The device of claim 1 wherein the first neural network is an embedded Gated Recurrent Unit (e-GRU) having weights trained to extract the features from the frequency domain audio signals for the second neural network.
3. The device of claim 2 wherein the second neural network is a second e-GRU neural network.
4. The device of claim 1 wherein the wake-up circuit is also coupled to the at least one microphone.
5. The device of claim 1 wherein the at least one microphone is a plurality of microphones.
6. The device of claim 1 wherein the candidate events comprise normal inhalation and exhalation breathing sounds, crackles, wheezes, coughs, and snoring.
7. The device of claim 6 wherein the candidate events further comprise gasping, choking, and speech sounds.
8. The device of claim 7 wherein the candidate events further comprise heart sounds.
9. The device of claim 8 being embedded within, or attached to, a pillow.
10. A system comprising:
the device configured for monitoring of physiological sounds of claim 1 ; and
a smart device configured to receive the uploaded candidate events and features from the digital radio, the smart device being selected from the group consisting of a body-area network (BAN) hub, a smartphone, a cellular-capable smartwatch, and a tablet computer.
11. The system of claim 10 further comprising code configured to classify and perform statistical analysis on the candidate events and features, the candidate events being classified into events comprising normal inhalation and exhalation breathing sounds, crackles, wheezes, coughs, and snoring.
12. The system of claim 11 wherein the first neural network is an embedded Gated Recurrent Unit (e-GRU) having weights trained to extract the features for the second neural network.
13. The device of claim 12 wherein the second neural network is a second e-GRU neural network.
14. The system of claim 11 wherein the smartphone is configured to relay the candidate events and features to a server, the server configured to classify and perform statistical analysis on the candidate events and features being executable on the server.
15. The device of claim 2 wherein the candidate events comprise normal inhalation and exhalation breathing sounds, crackles, wheezes, coughs, and snoring.
16. The device of claim 3 wherein the candidate events comprise normal inhalation and exhalation breathing sounds, crackles, wheezes, coughs, and snoring.
17. The device of claim 4 wherein the candidate events comprise normal inhalation and exhalation breathing sounds, crackles, wheezes, coughs, and snoring.
18. The device of claim 5 wherein the candidate events comprise normal inhalation and exhalation breathing sounds, crackles, wheezes, coughs, and snoring.