Methods and systems for processing data via an executable file on a monitor to reduce the dimensionality of the data and encrypting the data being transmitted over the wireless
Some embodiments include processing data via an executable file on a monitor to reduce the dimensionality of the data being transmitted over the wireless network. The output of the executable file also encrypts the data before being transmitted wireless to a remote server. The remote server receives the transmitted data and makes likelihood inferences based on the recorded data.
1. A monitor comprising:
a sensor configured to detect signals of a patient when engaged with a body of the patient; and
a signal processor configured to process detected patient signals through at least a first portion of a neural network to generate a first output, wherein the signal processor is local to the monitor;
wherein a computing system is configured to process the first output or a signal derived from the first output through at least a second portion of the neural network, wherein the computing system is external to the monitor.
2. The monitor of claim 1 , wherein the monitor is a wearable patch.
3. The monitor of claim 1 , wherein the monitor is a patch that is applied on the chest of the patient.
4. The monitor of claim 1 , wherein the monitor is an implantable device.
5. The monitor of claim 1 , wherein the first output of the first portion of the neural network is encrypted, wherein the computing system processes the encrypted first output through the second portion of the neural network.
6. The monitor of claim 1 , wherein the monitor includes a cardiac monitor, and wherein the patient signals are continuously recorded and are cardiac signals.
7. The monitor of claim 1 , wherein the monitor further comprises a patient trigger configured to depress and initiate recordation of an instance in time of a perceived cardiac event.
8. The monitor of claim 1 , wherein the signal processor is configured to select the neural network from a plurality of machine learning network based on a characteristic of the monitor.
9. The monitor of claim 8 , wherein the characteristic of the monitor comprises one or more of: a remaining amount of battery, a network characteristic between the monitor and the computing system, or a wear duration.
10. The monitor of claim 8 , wherein the characteristic includes one or more of: a characteristic of the patient or a severity of cardiac arrhythmia.
11. The monitor of claim 1 , wherein the signal processor is further configured to compress the first output, wherein the computing system is configured to decompress the compressed data; and wherein processing the first output through the second portion comprises processing the decompressed data.
12. The monitor of claim 1 , wherein the signal processor is further configured to quantize the first output of the first portion.
13. The monitor of claim 12 , wherein quantizing comprises rounding, truncating, or reducing a number of bits for the first output of the first portion.
14. The monitor of claim 12 , wherein the signal processor is further configured to determine an amount of quantization based on one or more of: a characteristic of the monitor or a lossless compression performance.
15. The monitor of claim 12 , wherein the signal processor is further configured to determine an amount of quantization based on at least one of: a processing power, a storage capacity, an amount of remaining storage capacity, or a network characteristic.
16. The monitor of claim 12 , wherein the signal processor is further configured to determine an amount of quantization based on an accuracy of the neural network.
17. The monitor of claim 1 , wherein the first portion of the neural network comprises an encoder, and the second portion of the neural network comprises a decoder.
18. The monitor of claim 17 , wherein the monitor includes a transmitter configured to transmit the first output of the encoder to the computing system.
19. The monitor of claim 1 , wherein the first portion of the neural network comprises a first subset of layers of the neural network, and the second portion of the neural network comprises a second subset of layers of the neural network.
20. The monitor of claim 19 , wherein the first subset of layers of the neural network and the second subset of layers of the neural network are trained together.
21. The monitor of claim 19 , wherein the dimensionality of the first output of the first subset of layers of the neural network is smaller than the data of the detected signals from the sensor.
22. The monitor of claim 19 , further comprising a receiver configured to receive an updated first subset of layers of the neural network from the computing system and updating the first subset of the layers of the neural network to the updated first subset of layers of the neural network, wherein the signal processor is further configured to process signals through the updated first subset of layers of the neural network.
23. The monitor of claim 19 , wherein the monitor further infers a likelihood of an occurrence of cardiac arrhythmia by processing the first output of the first subset through the second subset, wherein the first subset processes at least 24 hours of continuously detected, stored physiological signals.
24. The monitor of claim 1 , wherein to process the first output or a signal derived from the first output comprises determining or inferring a cardiac event of the patient.
25. The monitor of claim 24 , wherein the cardiac event comprises a likelihood of an occurrence of arrhythmia in the past of when the signals were detected.
26. The monitor of claim 24 , wherein the cardiac event comprises a likelihood of an occurrence of arrhythmia in the future of when the signals were detected.
27. The monitor of claim 24 , wherein the cardiac event comprises at least one of: a heart abnormality, a heart failure, a prognostic prediction, sleep apnea, hypertension, or blood pressure.
28. The monitor of claim 24 , wherein the cardiac event comprises at least one of: a condition only on electrocardiography (ECG) data.
29. The monitor of claim 24 , wherein the cardiac event comprises at least one of: a condition only on photoplethysmography (PPG) data.
30. The monitor of claim 24 , wherein the monitor comprises an accelerometer configured to gather motion data, wherein the computing system is configured to match motion data with the detected physiological signals to determine or infer the cardiac event.