Brain Activity Training Device Using Functional Connectivity Feedback

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current brain activity analysis and neurofeedback techniques for neurological/mental disorders lack practical applications, particularly in generating effective feedback for training and treating mental disorders like depression, as they do not clearly define how to extract brain activity characteristics for feedback during training.

Innovation Solution

A brain activity training apparatus and program that uses functional brain imaging to select specific brain connectivity patterns through machine learning, providing feedback based on the degree of approximation to a target pattern, allowing for training to change correlations among brain areas by calculating reward values and presenting them to the trainee.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If functional brain imaging is used to analyze brain activity patterns for neurofeedback training, then the ability to provide feedback for training mental disorders is improved, but the complexity of extracting and processing brain activity characteristics increases

Engineering Contradiction:
Improvefeedback effectiveness for trainingVSAvoidcomplexity of extracting brain activity characteristics
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts specific functional connectivity patterns from complex brain activity data by identifying and isolating characteristic connections between brain regions. Machine learning algorithms are used to extract relevant features from fMRI data, separating useful diagnostic information from noise and redundancy in the brain activity signals.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the brain into multiple functional regions and analyzes connectivity patterns between these segments. By dividing the brain activity data into discrete regional connections, the system simplifies the complex overall brain activity pattern into manageable components that can be systematically processed and feedback-based training can be applied.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If machine learning is used to select functional connectivity patterns, then the precision of identifying target brain patterns is improved, but the computational time and processing requirements increase

Engineering Contradiction:
Improveprecision of identifying target brain patternsVSAvoidcomputational processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary feature selection and pattern identification using machine learning algorithms during a training phase. The system pre-processes and analyzes brain activity data to identify and store characteristic functional connectivity patterns before actual diagnostic or treatment sessions, reducing real-time computational requirements.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent applies partial action by focusing machine learning analysis on specific functional connectivity patterns most relevant to the target brain function or disorder, rather than analyzing all possible brain connections. This selective approach maintains precision for identifying target patterns while reducing overall computational burden.

Inventive Principle:
Principle #16Partial or excessive action

3Ease of operation

If real-time neurofeedback is implemented, then the ability to provide immediate feedback for brain activity regulation is improved, but the imaging time required for analysis increases

Engineering Contradiction:
Improveimmediate feedback capabilityVSAvoidimaging analysis time
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The patent implements periodic action by providing feedback at regular intervals during the neurofeedback training session. Rather than continuous real-time feedback, the system analyzes brain activity patterns at periodic time points and provides feedback accordingly, maintaining training effectiveness while reducing the time required for each feedback cycle.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The patent uses preliminary action by pre-processing and preparing brain activity data analysis before the feedback is needed. The system has pre-established reference patterns and analysis protocols that allow rapid evaluation of current brain activity against these pre-prepared standards, reducing real-time analysis time while maintaining feedback immediacy.

Inventive Principle:
Principle #10Preliminary action

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

Enables effective feedback-based training to alter brain area connections, potentially improving mental health outcomes by aligning brain activity with healthy patterns, as demonstrated by improvements in depression symptom ratings.

Implementation Method 1

Functional brain imaging, such as functional Magnetic Resonance Imaging (fMRI), which visualizes hemodynamic reaction related to human brain activities using Magnetic Resonance Imaging (MRI)

Methodology Applied
Scientific EffectMagnetic Resonance Imaging:

Implementation Method 2

a disease determination system has been reported, which uses an NIRS (Near-InfraRed Spectroscopy) technique to classify mental disorders such as schizophrenia and depression based on features of hemoglobin signals measured by biological optical measurement

Methodology Applied
Scientific EffectNear-InfraRed Spectroscopy: Absorption Spectroscopy

Data Source

PatentEP3763288B1Brain activity training device, brain activity training method, and brain activity training program
Publication Date: 2024.06.19 ATR ADVANCED TELECOMM RES INST INT
  • EP3763288B1 patent drawingFigure 1
  • EP3763288B1 patent drawingFigure 2
  • EP3763288B1 patent drawingFigure 3

AI summary

Neurofeedback training executed by a brain activity training apparatus for performing a training to change the correlation of connectivity between brain areas utilizing correlation of measured connectivity between brain areas involves repetition of a plurality of trials. Each trial includes a resting period Trest, a self-regulation period TNF and a presenting period TScore presenting feedback information. The brain activity training apparatus calculates, from signals detected from a trainee in the resting period by a brain activity detecting device, a baseline level of degrees of activity of prescribed regions corresponding to the functional connectivity as the object of training, calculates, from signals detected in the self-regulation period and from the baseline level, time correlation of degrees of activity of the prescribed regions corresponding to the functional connectivity as the object of training, and calculates information to be fedback.