ML Artifact Rejection for TMS EEG Data

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Transcranial magnetic stimulation (TMS) electroencephalogram data is often contaminated with artifacts such as TMS stimuli, subject motion, and coil clicks, which skew analysis and prevent real-time guidance during procedures due to the inefficiency of manual artifact rejection.

Innovation Solution

A machine learning-based approach that decomposes and preprocesses TMS electroencephalogram data into independent components, using classifiers like Fisher linear discriminant analysis to identify and remove artefactual components, thereby generating clean data for real-time parameter adjustments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual artifact rejection is used to clean TMS electroencephalogram data, then measurement precision is improved, but productivity deteriorates due to inefficiency

Engineering Contradiction:

Improvedata cleanlinessVSAvoidprocessing speed

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces manual artifact rejection (mechanical human operation) with an automated machine learning system. The machine learning model automatically identifies and removes artefactual independent components from TMS electroencephalogram data, eliminating the need for manual processing while maintaining data cleanliness and significantly improving processing speed and productivity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If manual artifact rejection is used to clean TMS electroencephalogram data, then measurement precision is improved, but loss of time worsens

Engineering Contradiction:

Improvedata cleanlinessVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces time-consuming manual artifact rejection with an automated machine learning system that rapidly processes TMS electroencephalogram data. The machine learning model automatically decomposes data into independent components, identifies artefactual components, and removes them, dramatically reducing processing time while maintaining high data cleanliness standards.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent performs preliminary decomposition of TMS electroencephalogram data into independent components before the actual artifact rejection process. This preprocessing step organizes the data structure in advance, making the subsequent artifact identification and removal more efficient and faster, thereby reducing overall processing time while maintaining precision.

Inventive Principle:

Principle #10Preliminary action

3Productivity

If automated machine learning-based artifact rejection is implemented, then productivity is improved, but device complexity increases

Engineering Contradiction:

Improveprocessing efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent implements automated machine learning-based artifact rejection to replace manual processing, significantly improving productivity. While this introduces computational complexity, the complexity is managed through standardized machine learning pipelines and automated workflows, transforming human operational complexity into manageable computational processes that enhance overall system efficiency.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

4Productivity

If automated machine learning-based artifact rejection is implemented, then productivity is improved, but loss of time is reduced

Engineering Contradiction:

Improveprocessing efficiencyVSAvoidprocessing time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent replaces slow manual artifact rejection with automated machine learning processing, achieving simultaneous improvement in productivity and reduction in processing time. The automated system processes TMS electroencephalogram data much faster than manual methods while maintaining high processing efficiency through optimized computational algorithms.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent performs preliminary decomposition of electroencephalogram data into independent components and prepares feature vectors before the main artifact rejection process. This advance preparation optimizes the data structure and reduces computational burden during actual processing, enabling faster execution while maintaining high productivity through efficient workflow organization.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11577090B2Machine learning based artifact rejection for transcranial magnetic stimulation electroencephalogram

Publication Date: 2023.02.14 THE UNITED STATES OF AMERICA AS REPRESENTED BY THE DEPT OF VETERANS AFFAIRS

AI summary

A method for machine learning based artifact rejection is provided. The method may include applying a machine learning model to identify artefactual independent components in transcranial magnetic stimulation electroencephalogram data collected during a transcranial magnetic stimulation procedure. Clean transcranial magnetic stimulation electroencephalogram data is generated by removing, from the transcranial magnetic stimulation electroencephalogram data, the artefactual independent components. Real-time adjustments to parameters of the transcranial magnetic stimulation procedure may be performed based on the clean transcranial magnetic stimulation electroencephalogram data. Related systems and articles of manufacture, including computer program products, are also provided.