Attention Level Representation via EEG Feature Classification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for measuring attention levels in ADHD treatment using brain-computer interfaces (BCIs) rely on theta and beta wave activities, which are not directly correlated with attention, leading to users controlling wave activities without achieving higher attention levels.

Innovation Solution

A device and method that measure brain signals, extract temporal and spectral-spatial features, classify these features to differentiate concentration and non-concentration states, and combine scores to provide a direct representation of attention levels, presented through adaptive game controls.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If theta and beta wave activities are used to measure attention levels, then brain signals can be captured, but the measurement precision is poor because there is no direct correlation between wave activities and attention

Engineering Contradiction:
Improveattention level measurementVSAvoidcorrelation between measurement and attention
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent replaces the traditional theta/beta wave measurement approach with a machine learning-based classification system that processes multiple brain signal features (temporal, spectral, spatial) to directly predict attention levels. This substitution of the measurement methodology resolves the contradiction by establishing a direct computational link between brain signals and attention without relying on indirectly correlated wave activities.

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

Solution Approach 2:

The patent changes the measurement parameters from traditional theta/beta wave frequencies to multiple brain signal features including temporal features (mean, variance, skewness, kurtosis), spectral features (power spectral density), and spatial features (source localization). This parameter transformation enables direct attention measurement by selecting features that are actually correlated with attention states rather than using indirectly related wave activities.

Inventive Principle:
Principle #35Parameter changes

2Ease of operation

If users control theta and beta wave activities, then wave patterns can be modified, but attention levels do not improve because the control mechanism is not directly linked to attention

Engineering Contradiction:
Improvewave activity controlVSAvoidattention level improvement
Core Design Contradiction:
Ease of operationVSManufacturing precision

Solution Approach 1:

The patent implements a feedback mechanism where the attention level is continuously measured using machine learning classification, and this measurement is fed back to the user through visual or auditory cues. This direct feedback loop creates a closed-loop control system where users can immediately see the effect of their attention efforts, enabling precise control of attention rather than indirect control of wave activities.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system enables users to self-regulate their attention levels by providing real-time attention measurements and feedback. Users can independently monitor their own attention states and adjust their behavior accordingly without requiring external intervention or complex wave control mechanisms, making the attention control process intuitive and self-directed.

Inventive Principle:
Principle #25Self-service

3Measurement precision

If multiple brain signal features are extracted and classified, then attention measurement accuracy improves, but device complexity increases

Engineering Contradiction:
Improveattention level measurementVSAvoidsignal processing system
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the brain signal processing into distinct modules: temporal feature extraction, spectral feature extraction, spatial feature extraction, and machine learning classification. This modular segmentation allows each component to be optimized independently and simplifies the overall system architecture. The segmentation enables complex multi-feature analysis while maintaining manageable system complexity through structured organization of processing stages.

Inventive Principle:
Principle #1Segmentation

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 provides an accurate, quantitative, and continuous measure of attention levels, improving ADHD treatment by directly correlating brain signal analysis with attention, offering a more effective and user-centric training experience.

Implementation Method 1

BCI provides a direct communication pathway between a human brain and an external device. It relies on bio-signals such as electroencephalogram (EEG)

Methodology Applied
Scientific EffectElectroencephalogram (EEG):

Data Source

PatentEP2385789B1Device and method for generating a representation of a subject's attention level
Publication Date: 2019.03.13 NATIONAL UNIVERSITY OF SINGAPORE
  • EP2385789B1 patent drawingFigure 1~2
  • EP2385789B1 patent drawingFigure 3
  • EP2385789B1 patent drawingFigure 4

AI summary

A device and method for generating a representation of a subject's attention level. The device comprises means for measuring brain signals from the subject; means for extracting temporal features from the brain signals; means for classifying the extracted temporal features using a classifier to give a score x1; means for extracting spectral-spatial features from the brain signals; means for selecting spectral-spatial features containing discriminative information between concentration and non-concentration states from the set of extracted spectral-spatial features; means for classifying the selected spectral-spatial features using a classifier to give a score x2; means for combining the scores x1 and x2 to give a single score; and means for presenting said score to the subject.