Wrist-Worn Sensor Fusion for Real-Time Cognitive Load Classification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for assessing cognitive load are subjective, costly, or not suitable for real-time, continuous monitoring, and low-cost wearable sensors provide lower quality data, making accurate classification challenging.

Innovation Solution

A method and system using multi-modal wearable sensors to collect physiological signals, apply a multi-level feature extraction, select optimal features, augment training data, and use synthetic minority over-sampling to train a classification model for real-time cognitive load classification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If expensive physiological sensors are used for cognitive load assessment, then measurement precision is improved, but device cost increases

Engineering Contradiction:
Improvecognitive load measurement accuracyVSAvoidsensor system cost
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent employs low-cost wearable sensors (such as wrist-worn devices with skin conductance, heart rate, and temperature sensors) instead of expensive physiological sensors. These affordable sensors are sufficient when combined with multi-level feature extraction and machine learning models, achieving comparable or superior classification accuracy while dramatically reducing system cost and complexity.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent combines multiple low-cost wearable sensors (skin conductance, heart rate, respiration rate, skin temperature) to create a multi-modal sensing system. By merging data from these inexpensive sensors and applying multi-level feature extraction, the system achieves comprehensive cognitive load assessment without requiring any single expensive sensor.

Inventive Principle:
Principle #5Merging (Combining)

2Device complexity

If low cost wearable sensors are used for cognitive load assessment, then device cost is reduced, but measurement precision deteriorates

Engineering Contradiction:
Improvesensor system costVSAvoidcognitive load measurement accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent applies multi-level feature extraction that transforms raw sensor signals into multiple dimensions of information. Level 1 extracts basic signal features, Level 2 extracts temporal and spectral features, and Level 3 extracts non-linear dynamics features. This dimensional transformation creates rich feature representations from low-cost sensor data, enabling accurate cognitive load classification.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent employs multiple feature extraction methods (time-domain, frequency-domain, wavelet transform) that transform sensor parameters into different representations. By changing the parameter space and extracting features at multiple levels, the system maximizes the information content from low-cost sensor signals, improving measurement precision without increasing hardware cost.

Inventive Principle:
Principle #35Parameter changes

3Device complexity

If single physiological signal is used for cognitive load assessment, then device complexity is reduced, but measurement precision deteriorates

Engineering Contradiction:
Improvesensor system complexityVSAvoidcognitive load classification accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent combines four different physiological signals (skin conductance, heart rate, respiration rate, skin temperature) measured by multiple wearable sensors. This multi-modal approach captures complementary information about cognitive load, and the combined features achieve superior classification accuracy compared to any single signal alone.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent uses a unified machine learning classification model that processes features from multiple physiological domains (electrical, temporal, spectral, non-linear). This universal model handles diverse sensor types and feature representations, extracting meaningful patterns that indicate cognitive load states across all measured parameters.

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Ease of operation

If traditional self-report questionnaires are used for cognitive load assessment, then ease of operation is improved, but measurement precision deteriorates and real-time capability is lost

Engineering Contradiction:
Improveassessment method convenienceVSAvoidcognitive load measurement accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent replaces manual self-report questionnaires with an automated physiological monitoring system. Instead of requiring participants to verbally report their cognitive load, the system automatically measures physiological signals through wearable sensors and classifies cognitive load states using machine learning models, providing continuous objective monitoring.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent enables continuous real-time cognitive load assessment through wearable sensors that continuously monitor physiological signals. Unlike questionnaire-based methods that provide only snapshots at specific moments, the system continuously tracks cognitive load changes throughout the entire task duration, providing ongoing feedback.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentEP4109469B1Method and system for classification of cognitive load using data obtained from wearable sensors
Publication Date: 2025.10.22 TATA CONSULTANCY SERVICES LTD
  • EP4109469B1 patent drawingFigure 1
  • EP4109469B1 patent drawingFigure 2
  • EP4109469B1 patent drawingFigure 3A

AI summary

This disclosure relates generally to a method and system for classification of cognitive load (CL) using data obtained from wearable sensors. The disclosed method uses a multi-modal based approach using wrist-worn sensors for real time monitoring of CL in real world scenarios and improves the accuracy of detection of CL. A set of distinguishing features are selected from physiological signals received from the wrist-worn sensors. These features are used for training a classification model for classifying the CL of a patient into a no load or a high load. The set of distinguishing features are selected from domain specific features and signal property based generic features of the physiological signals. The disclosed method is used for classification of CL in scenarios such as to check how the cognitive load of a candidate varies during interviews, to assess the participants workload during online meetings and so on.