# US 11,298,071 B1 · Machine learning analytics in real time for health services

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,298,071 B1
- Document ID: US11298071B1
- Status: Granted
- Kind: B1
- Application number: 16/161,975
- Filed: 2018-10-16
- Granted: 2022-04-12
- Earliest claimed: 2017-10-16
- Assignees: University of South Florida
- Applicants: University of South Florida
- Inventors: Abolfazl Saghafi; Chris P. Tsokos
- Web page: https://patentgrep.com/patent/US11298071B1

## CPC

- A61B 5/4076: {Diagnosing or monitoring particular conditions of the nervous system}
- G06N 20/00: Machine learning
- A61B 5/1103: {Detecting muscular movement of the eye, e.g. eyelid movement}
- A61B 5/24: Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B 5/2415: {Measuring direct current [DC] or slowly varying biopotentials}
- A61B 5/316: Modalities, i.e. specific diagnostic methods
- A61B 5/374: Detecting the frequency distribution of signals, e.g. detecting delta, theta, alpha, beta or gamma waves
- A61B 5/4064: {Evaluating the brain (for intracranial pressure A61B5/031; for cerebral blood gases A61B5/14553; using EEG A61B5/369)}
- A61B 5/6803: {Head-worn items, e.g. helmets, masks, headphones or goggles}
- A61B 5/6814: {Head}
- A61B 5/725: {using specific filters therefor, e.g. Kalman or adaptive filters (specific diagnostics methods using using bioelectric or biomagnetic signals A61B5/316)}
- A61B 5/7253: {characterised by using transforms}
- A61B 5/7267: {involving training the classification device}
- A61B 5/7282: {Event detection, e.g. detecting unique waveforms indicative of a medical condition (cough events A61B5/0823; seizures A61B5/4094; sleep apnoea A61B5/4818)}
- G06F 17/18: for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)}
- G06N 7/00: Computing arrangements based on specific mathematical models
- A61B 5/251: Means for maintaining electrode contact with the body
- A61B 5/291: for electroencephalography [EEG]
- A61B 5/7203: {for noise prevention, reduction or removal}
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Methods and systems detect a random state change in a subject in real time. Eye state changes may be identified in encephalogram brain signals, or honeybee dance patterns may be classified. Multivariate signals including state change information are received via a plurality of channels. The signals are sampled and may be filtered to remove DC components. Statistical characteristics of the signals are monitored. When the statistical characteristics exceed a threshold during a critical time interval, a potential change of state is detected. The critical time segment of the signals may be filtered to generate respective state change artifact signals. The state change artifact signals are decomposed by MEMD, and intrinsic mode functions are generated. Features are extracted from the intrinsic mode functions. These steps may be repeated while the extracted features are provided to a logistic regression classifier that is used to predict a state of the subject.

## Claims

### Claim 1

A method for detecting a random state change in a physical subject, the method comprising:
receiving, using one or more electronic processors, a plurality of multivariate signals via a respective plurality of physical channels, wherein the plurality of multivariate signals comprise information about the state change in the physical subject, wherein the plurality of multivariate signals are sampled at a specified sampling rate;
monitoring, using the one or more electronic processors, values of statistical characteristics of the plurality of multivariate signals;
detecting, using the one or more electronic processors, a potential change of state in the physical subject when the monitored statistical characteristics values of at least one of the plurality of multivariate signals exceed a threshold;
low-pass filtering, using the one or more electronic processors, a time interval of the plurality of multivariate signals that precedes the detected potential change of state in the physical subject to generate a respective plurality of state change artifact signals;
simultaneously decomposing, using the one or more electronic processors, the plurality of state change artifact signals by multivariate empirical mode decomposition and generating a plurality of intrinsic mode functions;
extracting, using the one or more electronic processors, features from one or more of the plurality of intrinsic mode functions;
providing, using the one or more electronic processors, the extracted features to a logistic regression classifier;
repeating the steps of monitoring, detecting, low-pass filtering, decomposing, extracting, and providing the extracted features to the logistic regression classifier; and
predicting, using the one or more electronic processors, a state of the physical subject using the logistic regression classifier.

### Claim 2 (depends on 1)

The method of claim 1, wherein the plurality of multivariate signals are high pass filtered to remove DC components of the plurality of multivariate signals prior to the monitoring of the values of the statistical characteristics of the plurality of multivariate signals.

### Claim 3 (depends on 1)

The method of claim 1, wherein the plurality of multivariate signals comprise electroencephalogram (EEG) signals, and wherein the physical subject is an eye and potential states of the eye include an open state and a closed state.

### Claim 4 (depends on 1)

The method of claim 1, wherein the plurality of multivariate signals are nonlinear and non-stationary signals.

### Claim 5 (depends on 1)

The method of claim 1, wherein the statistical characteristics of the plurality of multivariate signals include maximum and minimum values of the plurality of multivariate signals.

### Claim 6 (depends on 1)

The method of claim 1, wherein the statistical characteristics of the plurality of multivariate signals are monitored at a reduced rate relative to the sampling rate.

### Claim 7 (depends on 1)

The method of claim 1, wherein the plurality of multivariate signals comprise electroencephalogram (EEG) signals, and wherein the respective plurality of state change artifact signals are eye opening or eye closing artifact signals.

### Claim 8 (depends on 1)

The method of claim 1, wherein the features are extracted from the one or more of the plurality of intrinsic mode functions over a reduced time interval relative to the time interval that precedes the detected potential change of state in the physical subject.

### Claim 9 (depends on 1)

The method of claim 1, wherein the features are extracted from the one or more of the plurality of intrinsic mode functions over a reduced time interval relative to the time interval that precedes the detected potential change of state in the physical subject, using a common spatial pattern method.

### Claim 10 (depends on 1)

The method of claim 1, wherein the features are extracted from the one or more of the plurality of intrinsic mode functions over a reduced time interval relative to the time interval that precedes the detected potential change of state in the physical subject, based on values of discriminant features of the one or more of the plurality of intrinsic mode functions.

### Claim 11

A system for detecting a random state change in a physical subject, the system comprising:
a memory storing program instructions; and
an electronic processor coupled to the memory, the electronic processor, through retrieval and execution of the program instructions, configured to:
receive a plurality of multivariate signals via a respective plurality of physical channels, wherein the plurality of multivariate signals comprise information about the state change in the physical subject, wherein the plurality of multivariate signals are sampled at a specified sampling rate;
monitor values of statistical characteristics of the plurality of multivariate signals;
detect a potential change of state in the physical subject when the monitored statistical characteristics values of at least one of the plurality of multivariate signals exceed a threshold;
low-pass filter a time interval of the plurality of multivariate signals that precedes the detected potential change of state in the physical subject to generate a respective plurality of state change artifact signals;
simultaneously decompose the plurality of state change artifact signals by multivariate empirical mode decomposition and generate a plurality of intrinsic mode functions;
extract features from one or more of the plurality of intrinsic mode functions;
provide the extracted features to a logistic regression classifier;
repeat the steps of monitoring, detecting, low-pass filtering, decomposing, extracting, and providing the extracted features to the logistic regression classifier; and
predict a state of the physical subject using the logistic regression classifier.

### Claim 12 (depends on 11)

The system of claim 11, wherein the plurality of multivariate signals are high pass filtered to remove DC components of the plurality of multivariate signals prior to the monitoring of the values of the statistical characteristics of the plurality of multivariate signals.

### Claim 13 (depends on 11)

The system of claim 11, wherein the plurality of multivariate signals comprise electroencephalogram (EEG) signals, and wherein the physical subject is an eye and potential states of the eye include an open state and a closed state.

### Claim 14 (depends on 11)

The system of claim 11 wherein the plurality of multivariate signals are nonlinear and non-stationary signals.

### Claim 15 (depends on 11)

The system of claim 11, wherein the statistical characteristics of the plurality of multivariate signals include maximum and minimum values of the plurality of multivariate signals.

### Claim 16 (depends on 11)

The system of claim 11, wherein the statistical characteristics of the plurality of multivariate signals are monitored at a reduced rate relative to the sampling rate.

### Claim 17 (depends on 11)

The system of claim 11, wherein the plurality of multivariate signals comprise electroencephalogram (EEG) signals, and wherein the respective plurality of state change artifact signals are eye opening or eye closing artifact signals.

### Claim 18 (depends on 11)

The system of claim 11, wherein the features are extracted from the one or more of the plurality of intrinsic mode functions over a reduced time interval relative to the time interval that precedes the detected potential change of state in the physical subject.

### Claim 19 (depends on 11)

The system of claim 11, wherein the features are extracted from the one or more of the plurality of intrinsic mode functions over a reduced time interval relative to the time interval that precedes the detected potential change of state in the physical subject, using a common spatial pattern method.

### Claim 20 (depends on 11)

The system of claim 11, wherein the features are extracted from the one or more of the plurality of intrinsic mode functions over a reduced time interval relative to the time interval that precedes the detected potential change of state in the physical subject, based on values of discriminant features of the one or more of the plurality of intrinsic mode functions.