EEG Correlation Analysis for ADHD Diagnostic Scoring

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods fail to effectively generate a score indicative of cognitive conditions like ADHD using EEG/MEG data, as they lack proper treatment of top-down attention signals and struggle to differentiate between stimulus-bounded surprise and uncertainty, leading to inefficiencies in diagnosing and assessing cognitive disorders.

Innovation Solution

An apparatus and method that calculate correlation values between specific time series from EEG/MEG data, focusing on source positions in the frontal lobe, ventral and dorsal attention networks, and ventral visual pathways to generate a score indicative of ADHD, leveraging the Shannonian brain model to reconcile surprise and uncertainty.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional EEG/MEG analysis methods are used to assess cognitive conditions, then the measurement process is simple, but the diagnostic accuracy and ability to distinguish ADHD from healthy subjects is insufficient

Engineering Contradiction:
Improvediagnostic accuracyVSAvoidanalysis complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the brain into specific functional regions (frontal lobe, ventral attention network, dorsal attention network, ventral visual pathway) and analyzes time series from each region separately. This segmentation allows targeted calculation of correlation values between specific brain networks, improving diagnostic accuracy for ADHD by focusing on relevant neural pathways rather than analyzing all brain regions uniformly.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transforms raw EEG/MEG signals into multiple derived parameters including time series, correlation values, and a composite score. By changing the parameter representation from raw signals to processed metrics that specifically capture attention network dynamics, the system achieves better differentiation between ADHD and healthy subjects while maintaining manageable analysis complexity.

Inventive Principle:
Principle #35Parameter changes

2Loss of information

If correlation values between all time series are calculated, then comprehensive brain network analysis is achieved, but the computational complexity and processing time increase significantly

Engineering Contradiction:
Improveinformation completenessVSAvoidprocessing time
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The patent extracts and focuses only on correlation values between time series from specific brain regions relevant to attention processing (frontal lobe, ventral and dorsal attention networks, ventral visual pathway). By extracting only the most diagnostically relevant correlations rather than calculating all possible pairwise correlations, the system maintains comprehensive analysis of attention networks while significantly reducing computational burden and processing time.

Inventive Principle:
Principle #2Taking out (Extraction)

3Adaptability or versatility

If prior art modeling approaches are used, then bottom-up visual attention is modeled, but top-down attention signals are not properly treated

Engineering Contradiction:
Improveattention model coverageVSAvoidmodeling accuracy
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent merges the analysis of both bottom-up and top-down attention mechanisms by simultaneously examining correlation values between time series from ventral visual pathways (bottom-up) and frontal lobe/attention network regions (top-down). This integration allows the model to capture the full spectrum of attention processing, improving both the coverage and reliability of cognitive condition assessment.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP4335370A1Apparatus for generating a score that is indicative for a cognitive condition like attention deficit hyperactivity disorder
Publication Date: 2024.03.13 BITSPHI DIAGNOSIS SL
  • EP4335370A1 patent drawingFigure 1~2
  • EP4335370A1 patent drawingFigure 3
  • EP4335370A1 patent drawingFigure 4

AI summary

The present disclosure relates to the generation of a score that is indicative for a subject having a cognitive condition, such as a cognitive disorder like attention deficit hyperactivity disorder, ADHD. In particular, the score may be generated based on electro- or magnetoencephalography (EEG/MEG) data by calculating one or more correlation values between pairs of time series that are determined based on the EEG/MEG data. An apparatus, system, computer-readable medium and computer-implemented method described herein receive EEG/MEG data of a subject. Then a first plurality of time series is determined based on the EEG/MEG data, wherein each of the first plurality of times series corresponds to a respective source position located inside a cranial cavity of the subject. Then, a first correlation value may be calculated for a first pair of time series, the first pair of time series being included in the determined first plurality of time series. A score is generated based on the first correlation value, the score being indicative for the subject having a cognitive condition, such as a cognitive disorder like attention deficit hyperactivity disorder, ADHD. The score is outputted, e.g., to a physician who may diagnose the subject.