EEG Cognitive Load Analysis via HHT Artifact Removal

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Low cost, low resolution EEG devices struggle to accurately measure cognitive load due to limited channels and noise contamination, leading to subjective placement of sensitive EEG channels and inadequate noise removal, which affects the accuracy of cognitive load determination.

Innovation Solution

A method using a Hilbert-Huang Transform filter to preprocess EEG signals from a low resolution EEG device with up to 14 channels, extracting Fast Fourier Transform based alpha and theta band power features, and classifying them using a supervised machine learning technique to determine cognitive load, specifically focusing on signals from the left-frontal brain lobe.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If low cost low resolution EEG devices are used, then cost is reduced, but measurement precision deteriorates due to fewer EEG channels and noise contamination

Engineering Contradiction:
ImprovecostVSAvoidcognitive load measurement accuracy
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent extracts and removes noise components from EEG signals using signal processing techniques. Specifically, it separates and eliminates artifacts such as eye movements, muscle activity, and electrical interference from the recorded EEG data, retaining only the relevant neural signals for cognitive load analysis

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces intermediate processing steps including signal filtering, artifact removal algorithms, and feature extraction methods that act as mediators between the raw EEG data and cognitive load measurement. These intermediate processes enhance the quality of data from low-resolution devices

Inventive Principle:
Principle #24Intermediary (Mediator)

2Device complexity

If low cost low resolution EEG devices with fewer channels are used, then device complexity is reduced, but reliability deteriorates due to missing sensitive EEG channels and inability to remove noise

Engineering Contradiction:
Improvenumber of EEG channelsVSAvoidcognitive load measurement reliability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent creates a virtual representation of additional EEG channels through signal processing and computational methods. By analyzing patterns from available channels and using algorithms to reconstruct or estimate signals that would be present with more channels, it compensates for the limited physical channel count

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent transforms the EEG data by changing parameters such as frequency domain representation, time-frequency analysis, and feature space transformation. These parameter changes enable reliable cognitive load measurement by emphasizing relevant signal characteristics while suppressing noise

Inventive Principle:
Principle #35Parameter changes

3Ease of operation

If standard EEG processing methods are used without specialized filtering, then processing simplicity is maintained, but measurement precision deteriorates due to noise contamination from non-cerebral artifacts

Engineering Contradiction:
Improveprocessing simplicityVSAvoidcognitive load measurement precision
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent applies preliminary signal processing and artifact removal techniques before main analysis. By pre-processing the EEG data to remove known artifacts and noise sources beforehand, it simplifies subsequent analysis while improving measurement precision

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent converts the challenge of noise contamination into an opportunity by using the noise characteristics themselves for identification and removal. By analyzing artifact patterns and using them as reference signals, the system effectively eliminates their harmful effects

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach effectively removes non-cerebral artifacts and reduces computational complexity, enabling accurate cognitive load measurement with improved separation between high and low cognitive load tasks, even with limited EEG channels, and enhances the accuracy of cognitive load determination.

Implementation Method 1

preprocessing, by the processor, the EEG signals using a Hilbert-Huang Transform (HHT) filter to remove a noise corresponding to one or more unrelated, non-cerebral artifacts

Methodology Applied
Scientific EffectHilbert-Huang Transform:

Implementation Method 2

extracting, by the processor, features comprising Fast Fourier Transform (FFT) based alpha and theta band power, from the preprocessed EEG signals

Methodology Applied
Scientific EffectFast Fourier Transform:

Data Source

PatentEP3011895B1Determining cognitive load of a subject from electroencephalography (EEG) signals
Publication Date: 2021.08.11 TATA CONSULTANCY SERVICES LTD
  • EP3011895B1 patent drawingFigure 1
  • EP3011895B1 patent drawingFigure 2
  • EP3011895B1 patent drawingFigure 3

AI summary

Disclosed is a method and system for determining a cognitive load of a subject from Electroencephalography (EEG) signals. EEG signals are received from EEG channels associated with a left-frontal brain lobe. EEG signals are associated with a subject performing cognitive task. EEG signals are received from a low resolution EEG device. EEG channels comprise four EEG channels associated with the left-frontal brain lobe. EEG signals are preprocessed using a Hilbert-Huang Transform (HHT) filter to remove a noise corresponding to one or more non-cerebral artifacts to generate preprocessed EEG signals. Features comprising Fast Fourier Transform (FFT) based alpha and theta band power are extracted from the preprocessed EEG signals. Feature vector is generated from the features. The feature vector is classified using a Support Vector Machine (SVM) classifier to determine the cognitive load of the subject.