Method and system for assessment of cognitive load from bio-potentials measured using wearable endosomatic device

Patent metadata

Field	Value
Publication number	US-11872040-B2
Application number	US-202017136899-A
Country	US
Kind code	B2
Filing date	Dec 29, 2020
Priority date	Jun 18, 2020
Publication date	Jan 16, 2024
Grant date	Jan 16, 2024

Abstract

Direct usage of endosomatic EDA has multiple challenges for practical cognitive load assessment. Embodiments of the method and system disclosed provide a solution to the technical challenges in the art by directly using the bio-potential signals to implement endosomatic approach for assessment of cognitive load. The method utilizes a multichannel wearable endosomatic device capable of acquiring and combining multiple bio-potentials, which are biomarkers of cognitive load experienced by a subject performing a cognitive task. Further, extracts information for classification of the cognitive load, from the acquired bio-signals using a set of statistical and a set of spectral features. Furthermore, utilizes a feature selection approach to identify a set of optimum features to train a Machine Learning (ML) based task classifier to classify the cognitive load experienced by a subject into high load task and low load task.

First claim

What is claimed is:

1. A processor implemented method for assessment of cognitive load from bio-potentials, comprising:
acquiring, via a multichannel wearable endosomatic device by one or more hardware processors, a plurality of bio-potential signals, wherein
the plurality of bio-potential signals provides a task specific data of a subject performing a cognitive task,
the cognitive task including one of a high load task or a low load task, and
the plurality of bio-potential signals is time stamped;
extracting, by the one or more hardware processors, baseline data and task data based on time stamps associated with each of the plurality of bio-potential signals, wherein
the baseline data corresponds to segment of each of the plurality of bio-potential signals when the subject experiences a nil cognitive load during the cognitive task being performed, and
the task data corresponds to segment of each of the plurality of bio-potential signals when the subject experiences an active cognitive load during the cognitive task being performed;

deriving, by the one or more hardware processors, a plurality of deviation signals from the plurality bio-potential signals by:
computing a mean of each of baseline data and the task data; and
subtracting the computed mean from each of the baseline data and the task data;

extracting, by the one or more hardware processors, a plurality of features from the plurality of deviation signals, wherein the plurality of features comprises a statistical set of features and a spectral set of features, wherein the spectral set of features comprises a mean frequency, a power at the mean frequency, a total power, a mean power, a median frequency, a power at median frequency, a peak frequency, a peak power, a signal band-power, a first moment, a second moment, and a third moment;
determining, by the one or more hardware processors, an optimum set of features from the plurality of features wherein the determined set of features best differentiates between the high load task or the low load task; and
training, by the one or more hardware processors, a classifier for classifying the cognitive task into one of the high load task or the low load task using the determined set of features, wherein the determined set of features is labeled as corresponding to the high load task or the low load task.

2. The processor implemented method of claim 1 , wherein the method comprises utilizing the trained classifier to classify the cognitive task performed by the subject in real time into the high load task and the low load task.

3. The processor implemented method of claim 1 , wherein the statistical set of features comprises:
a mean, median, a standard deviation, skewness, kurtosis, minimum levels and maximum levels of the plurality of bio-potential signals using the task data; and
normalized kurtosis and normalized variance values with respect to the baseline data and the task data.

4. A system for assessment of cognitive load from bio-potentials, comprising:
a memory storing instructions;
one or more Input/Output (I/O) interfaces; and
one or more hardware processors coupled to the memory via the one or more I/O interfaces, wherein the one or more hardware processors are configured by the instructions to:
acquire via a multichannel wearable endosomatic device, a plurality of bio-potential signals, wherein
the plurality of bio-potential signals provides a task specific data of a subject performing a cognitive task, and
the cognitive task including one or a high load task or a low load task, and
the plurality of bio-potential signals is time stamped;

extract baseline data and task data based on time stamps associated with each of the plurality of bio-potential signals, by the one or more hardware processors, wherein
the baseline data corresponds to segment of each of the plurality of bio-potential signals when the subject experiences a nil cognitive load during the cognitive task being performed, and
the task data corresponds to segment of each of the plurality of bio-potential signals when the subject experiences an active cognitive load during the cognitive task being performed;
derive a plurality of deviation signals from the plurality bio-potential signals by:
computing mean of each of baseline data and the task data; and
subtracting the computed mean from each of the baseline data and the task data;

extract a plurality of features from the plurality of deviation signals, wherein the plurality of features comprising a statistical set of features and a spectral set of features, wherein the spectral set of features comprises a mean frequency, a power at the mean frequency, a total power, a mean power, a median frequency, a power at median frequency, a peak frequency, a peak power, a signal band-power, a first moment, a second moment, and a third moment;
determine an optimum set of features from the plurality of features wherein the determined set of features best differentiates between the high load task or the low load task; and
train a classifier for classifying the cognitive task into one of the high load task and the low load task using the determined set of features wherein the determined set of features is labeled as corresponding to the high load task or the low load task.

5. The system of claim 4 , wherein the one or more hardware processors are further configured to utilize the trained classifier to classify the cognitive task performed by the subject in real time into the high load task and the low load task.

6. The system of claim 4 , wherein the set of statistical features comprises:
mean, median, standard deviation, skewness, kurtosis, minimum levels and maximum levels of the plurality of bio-potential signals using the task data; and
normalized kurtosis and normalized variance values with respect to the baseline data and the task data.

7. One or more non-transitory machine-readable information storage mediums comprising one or more instructions, which when executed by one or more hardware processors causes the one or more hardware processors to perform the steps of:
acquiring, via a multichannel wearable endosomatic device, a plurality of bio-potential signals, wherein
the plurality of bio-potential signals provides a task specific data of a subject performing a cognitive task,
the cognitive task including one of a high load task or a low load task, and
the plurality of bio-potential signals is time stamped;

extracting baseline data and task data based on time stamps associated with each of the plurality of bio-potential signals, wherein
the baseline data corresponds to segment of each of the plurality of bio-potential signals when the subject experiences a nil cognitive load during the cognitive task being performed, and
the task data corresponds to segment of each of the plurality of bio-potential signals when the subject experiences an active cognitive load during the cognitive task being performed;

deriving a plurality of deviation signals from the plurality bio-potential signals by:
computing a mean of each of the baseline data and the task data; and
subtracting the computed mean from each of the baseline data and the task data;

extracting a plurality of features from the plurality of deviation signals, wherein the plurality of features comprises a statistical set of features and a spectral set of features, wherein the spectral set of features comprises a mean frequency, a power at the mean frequency, a total power, a mean power, a median frequency, a power at median frequency, a peak frequency, a peak power, a signal band-power, a first moment, a second moment, and a third moment;
determining an optimum set of features from the plurality of features wherein the determined set of features best

Assignees

• Tata Consultancy Services Ltd

Inventors

• Chatterjee Debatri
• Jaiswal Dibyanshu
• Pal Arpan
• Ramakrishnan Ramesh Kumar
• Ghosh Ratna
• Moulick Madhurima
• Ranjan Rajesh

Classifications

• A61B5/24

  Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof · CPC title

• A61B5/165 Primary

  Evaluating the state of mind, e.g. depression, anxiety · CPC title

• A61B5/0533

  Measuring galvanic skin response · CPC title

• A61B5/7267

  involving training the classification device · CPC title

• A61B5/7278

  Artificial waveform generation or derivation, e.g. synthesizing signals from measured signals · CPC title