Method and system for determining inactive state and its implication over cognitive load computation of a person

Patent metadata

Field	Value
Publication number	US-11076796-B2
Application number	US-201715726559-A
Country	US
Kind code	B2
Filing date	Oct 6, 2017
Priority date	Oct 7, 2016
Publication date	Aug 3, 2021
Grant date	Aug 3, 2021

Abstract

A method and system for determining cognitive load of a person using a modified baseline is provided. The person is asked to perform a series of activities including staying in eye closed rest state and baseline state and performing a trial state. Simultaneously, EEG signal and GSR signal of the person are captured. The EEG signal and the GSR signal are preprocessed and segmented. The EEG and GSR signals are then used to determine a first set and a second set of inactive states from the baseline interval and the rest interval. The most inactive window is then identified out of the first set of inactive states. The most inactive window is determined from the rest interval of the person. The inactive window is used as the modified baseline to measure the cognitive load of the person.

First claim

What is claimed is:

1. A method for determining a cognitive load of a person using a modified baseline, the method comprising a processor implemented steps of:
instructing the person to stay in an eyes closed rest state, stay in a baseline state and carry out a trial state respectively, wherein, a task is performed by the person during the trial state;
sensing an Electroencephalogram (EEG) signal of the person using an EEG sensor;
sensing a Galvanic Skin Response (GSR) signal of the person using a GSR sensor;
preprocessing the EEG signal and the GSR signal using a preprocessor;
segmenting the preprocessed EEG signal and the preprocessed GSR signal into a rest interval corresponding to the rest state, a baseline interval corresponding to the baseline state and a trial interval corresponding to the trial state;
performing power spectral analysis of the EEG signal to determine a first set of inactive states of the person from the baseline interval and the rest interval;
identifying a most inactive window out of the first set of inactive states, wherein the most inactive window is during the rest interval of the person;
determining the cognitive load of the person using the EEG signal of the person during the trial state and by using the most inactive window as the modified baseline; and
validating the identification of the first inactive window, wherein validating comprises:
performing power spectral analysis of the GSR signal to determine a set of second inactive states of the person from the baseline interval and the rest interval;
identifying a most inactive window out of the set of second inactive states, wherein the most inactive window is during the rest interval of the person; and
validating the most inactive state derived from the EEG signal with respect to the most inactive state derived from the GSR signal.

2. The method of claim 1 , wherein the task includes a high load task and a low load task.

3. The method of claim 1 further comprising validating the high load task and the low load task, using the most inactive windows derived from the EEG signal and the GSR signal.

4. The method of claim 1 further comprises of sub dividing the preprocessed EEG signal and the preprocessed GSR signal into windows of 1 second duration.

5. The method of claim 1 , wherein the step of determining the cognitive load further comprises:
decomposing the EEG signal into a time frequency domain using S Transform;
calculating a mean frequency and power of the decomposed EEG signal at various frequencies;
calculating change in product of the mean frequency and the power in both alpha range of 8-12 Hz and theta range of 4-7 Hz of the EEG signal between the stimulus interval and the rest interval; and
calculating the cognitive load using the change in the product of the mean frequency and the power in each of the windows.

6. The method of claim 1 , wherein the power spectral analysis of the EEG signal further comprises:
determining energies of a plurality of frequency bands of the EEG signal using Fast Fourier Transform;
creating a metric of energies to validate in the rest state, conventional baseline state and the trial state, wherein the metric comprises of presence of inactivity and presence of activity in the window; and
calculating an activity measure corresponding to the activity state and an inactivity measure corresponding to the inactivity state of the person.

7. The method of claim 1 , wherein the power spectral analysis of the GSR signal further comprises extraction of tonic component and phasic component of the GSR signal, wherein the tonic component is indicative of the inactive state of the person.

8. The method of claim 1 , wherein in the rest state of the person is resting with eyes closed.

9. A system for determining a cognitive load of a person using a modified baseline, the system comprises:
an input/output interface for instructing the person to stay in an eyes closed rest state, stay in a baseline state and carry out a trial state respectively, wherein, a task is performed by the person during the trial state;
an Electroencephalogram (EEG) sensor for sensing an EEG signal of the person;
a Galvanic Skin Response (GSR) sensor for sensing a GSR signal of the person;
a preprocessor for preprocessing the EEG signal and the GSR signal;
a memory; and
a processor coupled with the memory, wherein the processor further configured to perform the steps of:
segmenting the preprocessed EEG signal and the preprocessed GSR signal into a rest interval corresponding to the rest state, a baseline interval corresponding to the baseline state and a trial interval corresponding to the trial state;
performing power spectral analysis of the EEG signal to determine a first set of inactive states of the person from the baseline interval and the rest interval;
identifying a most inactive window out of the first set of inactive states, wherein the most inactive window is during the rest interval of the person;
determining the cognitive load of the person using the EEG signal of the person during the trial state and by using the most inactive window as the modified baseline; and
validating the identification of the first inactive window, wherein validating comprises:
performing power spectral analysis of the GSR signal to determine a set of second inactive states of the person from the baseline interval and the rest interval;
identifying a most inactive window out of the set of second inactive states, wherein the most inactive window is during the rest interval of the person; and
validating the most inactive state derived from the EEG signal with respect to the most inactive state derived from the GSR signal.

10. A non-transitory computer readable medium embodying a program executable in a computing device for determining a cognitive load of a person using a modified baseline, the program comprising a processor implemented steps of:
instructing the person to stay in an eyes closed rest state, stay in a baseline state and carry out a trial state respectively, wherein, a task is performed by the person during the trial state;
sensing an Electroencephalogram (EEG) signal of the person using an EEG sensor;
sensing a Galvanic Skin Response (GSR) signal of the person using a GSR sensor;
preprocessing the EEG signal and the GSR signal using a preprocessor;
segmenting the preprocessed EEG signal and the preprocessed GSR signal into a rest interval corresponding to the rest state, a baseline interval corresponding to the baseline state and a trial interval corresponding to the trial state;
performing power spectral analysis of the EEG signal to determine a first set of inactive states of the person from the baseline interval and the rest interval;
identifying a most inactive window out of the first set of inactive states, wherein the most inactive window is during the rest interval of the person;
determining the cognitive load of the person using the EEG signal of the person during the trial state and by using the most inactive window as the modified baseline; and
validating the identification of the first inactive window, wherein validating comprises:
performing power spectral analysis of the GSR signal to determine a set of second inactive states of the person from the baseline interval and the rest interval;
identifying a most inactive window out of the set of second inactive states, wherein the most inactive window is during the rest interval of the person; and
validating the most inactive state derived from the EEG signal with respect to the most inactive state derived from the GSR signal.

Assignees

• Tata Consultancy Services Ltd
• Tata Consultancy Services Llc

Inventors

• Gavas Rahul Dasharath
• Das Rajat Kumar
• Das Pratyusha
• Chatterjee Debatri
• Sinha Aniruddha

Classifications

• A61B5/372

  Analysis of electroencephalograms · CPC title

• A61B5/165 Primary

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

• A61B5/374

  Detecting the frequency distribution of signals, e.g. detecting delta, theta, alpha, beta or gamma waves · CPC title

• A61B5/0533 Primary

  Measuring galvanic skin response · CPC title

• A61B5/7264

  Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems · CPC title