Energy-efficient on-chip classifier for detecting physiological conditions

Patent metadata

Field	Value
Publication number	US-2020388397-A1
Application number	US-202016946151-A
Country	US
Kind code	A1
Filing date	Jun 8, 2020
Priority date	Jun 7, 2019
Publication date	Dec 10, 2020
Grant date	—

Abstract

Methods, systems, and devices are disclosed for an efficient hardware architecture to implement gradient boosted trees for detecting biological conditions. For example, a method of detecting a biological condition includes receiving, by a device, a plurality of physiological signals from a plurality of input channels of the device, selecting, based on a trained prediction model, one or more input channels from the plurality of input channels, converting the one or more physiological signals received from the one or more input channels to one or more digital physiological signals, identifying, by using the plurality of gradient boosted decision trees, the selected characteristic in the one or more digital physiological signals, and determining a presence of a physiological condition based on an addition of the output values obtained from the plurality of gradient boosted decision trees.

First claim

What is claimed is:

1 . A method of detecting a biological condition, comprising:
receiving, by a device, a plurality of physiological signals from a plurality of input channels of the device; selecting, based on a trained prediction model, one or more input channels from the plurality of input channels,
wherein the trained prediction model indicates the one or more input channels and configurations of a plurality of gradient boosted decision trees for identification of a selected characteristic of one or more physiological signals from the plurality of physiological signals;
converting the one or more physiological signals received from the one or more input channels to one or more digital physiological signals; identifying, by using the plurality of gradient boosted decision trees, the selected characteristic in the one or more digital physiological signals,
wherein the identifying the selected characteristic includes providing output values by the plurality of gradient boosted decision trees; and
determining a presence of a physiological condition based on an addition of the output values obtained from the plurality of gradient boosted decision trees.

2 . The method of claim 1 ,
wherein the plurality of gradient boosted decision trees operate in parallel, wherein the identifying the characteristic is performed within an optimum time that is determined based on a plurality of times associated with the plurality of gradient boosted decision trees, and wherein each of the plurality of times indicate an amount of time associated with obtaining an output value from an associated gradient boosted decision tree.

3 . The method of claim 1 , wherein each gradient boosted decision tree is associated with a programmable finite impulse response (FIR) filter that filters or bypasses a digital physiological signal based on the selected characteristic.

4 . The method of claim 3 ,
wherein the device includes a memory that stores the plurality of gradient boosted decision trees and coefficient values for the programmable FIR filter for each gradient boosted decision tree, and wherein the coefficient values are based on the selected characteristics.

5 . The method of claim 3 ,
wherein the programmable FIR filter includes a first stage that outputs a downsampled physiological signal that is obtained by downsampling the digital physiological signal, wherein the programmable FIR filter includes a second stage that includes a tunable bandpass filter that filters the downsampled physiological signal, and wherein bandwidth related parameters of the tunable bandpass filter are determined based on the selected characteristic.

6 . The method of claim 5 , wherein any one or more of the first stage and the second stage are bypassed based on the selected characteristic.

7 . The method of claim 1 , wherein the selecting the one or more input channels is performed using a multiplexer associated with each of the plurality of gradient boosted decision trees.

8 . The method of claim 1 , wherein the selecting, the converting, the identifying, and the determining is performed for the one or more input channels that are selected without buffering data from the plurality of input channels other than the one or more input channels.

9 . The method of claim 1 ,
wherein a number of the plurality of gradient boosted decision trees is up to eight, and wherein each gradient boosted decision tree has a maximum pre-determined depth of four.

10 . A device, comprising:
a processor configured to: receive a plurality of physiological signals from a plurality of input channels; select, based on a trained prediction model, one or more input channels from the plurality of input channels,
wherein the trained prediction model indicates the one or more input channels and configurations of a plurality of gradient boosted decision trees for identification of a selected characteristic of one or more physiological signals from the plurality of physiological signals;
convert the one or more physiological signals received from the one or more input channels to one or more digital physiological signals; identify, by using the plurality of gradient boosted decision trees, the selected characteristic in the one or more digital physiological signals,
wherein the identifying the selected characteristic includes providing output values by the plurality of gradient boosted decision trees; and
determine a presence of a physiological condition based on an addition of the output values obtained from the plurality of gradient boosted decision trees.

11 . The device of claim 10 ,
wherein the plurality of gradient boosted decision trees are configured to operate in parallel, wherein the processor is configured to identify the characteristic within an optimum time that is determined based on a plurality of times associated with the plurality of gradient boosted decision trees, and wherein each of the plurality of times indicate an amount of time associated with obtaining an output value from an associated gradient boosted decision tree.

12 . The device of claim 10 , wherein each gradient boosted decision tree is associated with a programmable finite impulse response (FIR) filter that is configured to filter or bypass a digital physiological signal based on the selected characteristic.

13 . The device of claim 12 ,
wherein the device includes a memory that is configured to store the plurality of gradient boosted decision trees and coefficient values for the programmable FIR filter for each gradient boosted decision tree, and wherein the coefficient values are based on the selected characteristics.

14 . The device of claim 12 ,
wherein the programmable FIR filter includes a first stage that is configured to output a downsampled physiological signal obtained by downsampling the digital physiological signal, wherein the programmable FIR filter includes a second stage that includes a tunable bandpass filter that is configured to filter the downsampled physiological signal, and wherein bandwidth related parameters of the tunable bandpass filter are determined based on the selected characteristic.

15 . The device of claim 14 , wherein any one or more of the first stage and the second stage are bypassed based on the selected characteristic.

16 . A non-transitory machine-readable medium having machine executable instructions stored thereon that, when executed by one or more processors, direct the one or more processors to perform a method comprising:
receiving, by a device, a plurality of physiological signals from a plurality of input channels of the device; selecting, based on a trained prediction model, one or more input channels from the plurality of input channels,
wherein the trained prediction model indicates the one or more input channels and configurations of a plurality of gradient boosted decision trees for identification of a selected characteristic of one or more physiological signals from the plurality of physiological signals;
converting the one or more physiological signals received from the one or more input channels to one or more digital physiological signals; identifying, by using the plurality of gradient boosted decision trees, the selected characteristic in the one or more digital physiological signals,
wherein the identifying the selected characteristic includes providing output values by the plurality of gradient boosted decision trees; and
determining a presence of a physiological condition based on an addition of the output values obtained from the plurality o

Assignees

• Univ Cornell

Inventors

• Shoaran Mahsa
• Taghavi Milad
• Haghl Benyamin A
• Farivar Masoud
• Emami Azita

Classifications

• G06N3/045

  Combinations of networks · CPC title

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/084

  Backpropagation, e.g. using gradient descent · CPC title

• G06N20/10

  using kernel methods, e.g. support vector machines [SVM] · CPC title