EEG Cognitive Load Evaluation via Unsupervised Clustering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods for measuring cognitive load using EEG signals face challenges such as variability in signal intensity and range due to spatial shifts in electrode positioning, requiring time-consuming supervised learning approaches that involve two-stage user training and fail to address normalization issues.

Innovation Solution

The method involves receiving EEG data, splitting it into slots, extracting features in both frequency and time domains, and grouping data points into clusters using unsupervised learning techniques like Fuzzy c-Means or Component-wise Fuzzy c-Means algorithms to evaluate cognitive load levels.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If supervised learning approach is used for cognitive load measurement, then classification accuracy can be achieved, but user training time and system complexity increase due to two-stage training requirement

Engineering Contradiction:
Improvecognitive load measurement accuracyVSAvoiduser training time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs self-calibration by automatically determining normalization parameters from the user's own EEG data without requiring external reference data or manual training. The unsupervised learning algorithm autonomously identifies cognitive load levels and establishes normalization factors, eliminating the need for two-stage supervised training while maintaining measurement accuracy

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system changes the learning approach from supervised to unsupervised learning, transforming the training requirement from manual two-stage training to automatic parameter estimation. This parameter change in the learning methodology eliminates training time requirements while preserving measurement precision through automated normalization

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If supervised learning approach is used for cognitive load measurement, then classification can be performed, but device complexity increases due to two-step model generation requirement

Engineering Contradiction:
Improvecognitive load classification accuracyVSAvoidmodel generation complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system eliminates the need for complex two-step model generation by using unsupervised learning that automatically adapts to each user's EEG characteristics. The algorithm self-calibrates normalization parameters and generates classification models independently, reducing system complexity while maintaining classification accuracy

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

Instead of requiring manual training data and supervised learning models, the system inverts the approach by using unsupervised learning that discovers patterns automatically from raw EEG data. This inversion of the learning paradigm simplifies the system architecture by removing the need for pre-trained models and manual training procedures

Inventive Principle:
Principle #13The other way round (Inversion)

3Productivity

If EEG signals are not normalized, then raw signal processing can be performed, but measurement accuracy deteriorates due to variability in signal intensity and range from spatial shifts

Engineering Contradiction:
Improvesignal processing efficiencyVSAvoidcognitive load measurement accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system performs automatic self-normalization by computing normalization parameters from the user's own EEG signal characteristics. The unsupervised learning algorithm automatically adjusts for spatial shifts and intensity variations without requiring manual calibration or external reference signals, maintaining both processing efficiency and measurement accuracy

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system dynamically changes the signal parameters by computing normalized EEG values based on individual user characteristics. This parameter transformation adjusts the signal intensity and range to compensate for spatial shifts while preserving the temporal structure and cognitive load information for accurate measurement

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3047475B1System and method for evaluating a cognitive load on a user corresponding to a stimulus
Publication Date: 2021.08.18 TATA CONSULTANCY SERVICES LTD
  • EP3047475B1 patent drawingFigure 1
  • EP3047475B1 patent drawingFigure 2
  • EP3047475B1 patent drawingFigure 3A~3B

AI summary

System and method for evaluating a cognitive load on a user, corresponding to a stimulus is disclosed. Electroencephalogram (EEG) data corresponding to the stimulus of a user is received. The stimulus corresponds to a mental task performed by the user. The EEG data is split into a plurality of slots. A slot of the plurality of slots comprises a subset of the EEG data. One or more EEG features are extracted from the subset of the EEG data. The one or more EEG features are represented in one of a frequency domain and a time domain. A plurality of data points present in the one or more EEG features is grouped into two or more clusters using an unsupervised learning technique. The two or more clusters comprise one or more data points of the plurality of data points. The one or more data points correspond to a level of the cognitive load.