Brain Activity Training Using Discriminator Feedback

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current brain activity analysis using functional brain imaging and neurofeedback techniques are not yet practical for treating neurological and mental disorders, particularly due to the lack of established biomarkers for predicting diagnostic results and developing therapeutic agents, and the need for more effective training methods to achieve desirable brain states.

Innovation Solution

A brain activity training apparatus and system that utilizes correlations among brain regions measured by functional brain imaging as feedback information, employing a discriminator generated from pre-measured signals to calculate reward values and present them to subjects, enabling training to change these correlations and potentially treat neurological and mental disorders.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If functional brain imaging is used to measure brain activities, then brain activity patterns can be detected, but the lack of established biomarkers prevents practical application for treating neurological and mental disorders

Engineering Contradiction:
Improvebrain activity detectionVSAvoiddiagnostic prediction accuracy
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The system performs preliminary measurement of brain activities from multiple second subjects to generate a discriminator before actual training. This pre-computation of the discriminator (which encodes target brain state patterns) enables subsequent real-time feedback to reliably guide subjects toward desirable brain states, resolving the lack of established biomarkers for treatment

Inventive Principle:
Principle #10Preliminary action

2Ease of operation

If real-time neurofeedback is implemented, then subjects can receive feedback about brain activities, but the complexity of generating effective feedback signals from multiple brain regions remains unresolved

Engineering Contradiction:
Improveneurofeedback implementationVSAvoidfeedback signal processing
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The system extracts a condensed discriminator signal from complex multi-region brain activity patterns of multiple second subjects. This extracted discriminator serves as a simplified feedback target that captures essential brain state information, making real-time neurofeedback practical while avoiding the complexity of processing individual region correlations directly

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The discriminator acts as an intermediary between raw brain activity measurements and feedback presentation to subjects. It translates complex multi-dimensional brain correlation data into a manageable feedback signal that can be processed in real-time and presented to guide subjects toward target brain states

Inventive Principle:
Principle #24Intermediary (Mediator)

3Manufacturing precision

If correlations of brain activities from multiple subjects are used to generate a discriminator, then target brain states can be defined, but the computational complexity increases

Engineering Contradiction:
Improvetarget brain state definitionVSAvoiddiscriminator generation process
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The system merges brain activity data from multiple second subjects to generate a consensus discriminator. By combining measurements across multiple subjects, the system defines more robust target brain states that represent desirable patterns, while the merging process itself averages out individual variations and reduces computational burden compared to processing each subject separately

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP2992823B1Brain activity training device
Publication Date: 2021.06.30 ATR ADVANCED TELECOMM RES INST INT
  • EP2992823B1 patent drawingFigure 1
  • EP2992823B1 patent drawingFigure 2
  • EP2992823B1 patent drawingFigure 3

AI summary

Provided is a brain activity training apparatus for training to cause a change in correlation of connectivity among brain regions, utilizing measured correlations of connections among brains regions as feedback information. From measured data of resting-state functional connectivity MRI of a healthy group and a patient group (S102), correlation matrix of degree of brain activities among prescribed brain regions is derived for each subject. Feature extraction is executed (S104) by regularized canonical correlation analysis on the correlation matrix and attributes of the subject including a disease/healthy label of the subject. Based on the result of regularized canonical correlation analysis, by discriminant analysis through sparse logistic regression, a discriminator is generated (S108). The brain activity training apparatus feeds back a reward value to the subject based on the result of discriminator on the data of functional connectivity MRI of the subject.