EEG Classification for Perception Quantification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for determining a person's perception ability are often inconclusive due to individual differences in EEG data, making it difficult to quantify cognition effectively and independently of motor actions.

Innovation Solution

A method involving the application of different types of stimuli, such as tones, vibrations, or visual stimuli, with corresponding mental activities, where EEG data is recorded and analyzed to determine the distinguishability between stimuli, using classification analysis to quantify perception ability independently of motor actions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If classification methods based on existing measured values are used, then the determination of perception ability can be performed, but the results are inconclusive due to individual differences between subjects

Engineering Contradiction:
Improveperception ability determinationVSAvoidmeasurement consistency
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent transforms the classification approach from using fixed existing measured values to using dynamically determined classifications based on distinguishability measures. The classification is adapted to each subject's individual EEG characteristics by determining how well EEG data associated with specific stimuli can be distinguished from other stimuli, thereby accounting for individual differences while maintaining measurement reliability

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent performs preliminary classification analysis to determine the distinguishability measure before final perception ability quantification. By pre-determining the classification that maximizes distinguishability between different stimulus types for each individual subject, the system establishes a subject-specific baseline that improves subsequent measurement precision and reliability

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If existing EEG classification methods are used, then perception determination can be performed, but the method is complex and time-consuming

Engineering Contradiction:
Improvecognition determinationVSAvoidtest duration
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts only the essential feature needed for perception determination - the distinguishability measure of EEG data associated with different stimuli. By focusing specifically on whether EEG responses to different stimulus types can be distinguished from each other, rather than performing comprehensive EEG analysis, the method achieves accurate cognition determination with reduced time and complexity

Inventive Principle:
Principle #2Taking out (Extraction)

3Ease of operation

If motor actions are required for testing, then communication ability can be assessed, but the method cannot distinguish between perception ability and motor execution ability

Engineering Contradiction:
Improvecommunication assessmentVSAvoidperception ability measurement
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent replaces the mechanical response system (motor actions) with a neural signal detection system (EEG classification). Instead of requiring subjects to physically respond to stimuli, the system detects and classifies neural responses directly from EEG data, substituting mechanical observation with electrical signal analysis. This eliminates the confounding factor of motor execution ability while maintaining ease of operation for communication assessment

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

Data Source

PatentEP3060112B1Method for quantifying the perceptive faculty of a person
Publication Date: 2019.11.20 GUGER CHRISTOPH
  • EP3060112B1 patent drawingFigure 1
  • EP3060112B1 patent drawingFigure 2
  • EP3060112B1 patent drawingFigure 3

AI summary

The invention relates to a method for quantifying the perceptive faculty of a person (1), wherein a set of at least two possible differently perceptible types of stimuli (S) that can be applied to the test subject (1) is provided, and the test subject (1) is set intellectual tasks which, in the presence of a stimulus (S), are intended to be performed according to the nature of this stimulus (S), wherein a plurality of test steps is carried out, wherein for each of the test steps a particular type of stimulus (S) is chosen from the set of possible types of stimuli (S), in particular according to random criteria, a stimulus (S) of the respectively chosen type of stimuli is applied to the person (1), and, within a time range before, during or after the application of the respective stimulus (S), EEG data of the person are determined and recorded, wherein the time range preferably has a duration of 1 to 10 seconds, and the respectively determined EEG data or data derived therefrom are assigned to the respective type of stimulus (S), wherein classification analysis (105) is used to determine a measure (M) of whether the EEG data assigned to a defined stimulus (S) are distinguishable from the EEG data assigned to a stimulus (S) of a different type, and wherein the measure (M) of the distinguishability of the EEG data of different stimuli (S) is used as a measure of the perceptive faculty.