TMS Prediction System Using EEG Brain Network Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current treatments for neuropsychiatric conditions like ADHD lack effective prediction methods for treatment outcome, leading to inefficiencies and intolerance in many patients.

Innovation Solution

A system combining transcranial magnetic stimulation (TMS) with electrophysiological signal detection using EEG, where a computer processor applies TMS pulses, detects signals, and predicts treatment outcomes by analyzing frequency bands and brain network activity patterns.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional treatment methods for neuropsychiatric conditions are used, then treatment can be applied to patients, but there is no effective prediction method for treatment outcome leading to inefficiencies and intolerance

Engineering Contradiction:
Improvetreatment outcome prediction accuracyVSAvoidlack of predictive information
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The system performs preliminary assessment of brain electrical activity patterns before treatment to predict treatment outcome. By analyzing EEG spectra and brain network activity in advance, the system determines whether a patient is likely to respond to TMS therapy, enabling preliminary classification of treatment responders vs. non-responders before actual treatment begins.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from measured brain electrical activity patterns to adjust and optimize treatment predictions. By continuously monitoring EEG spectra, frequency bands, and brain network activity, the system refines its prediction of treatment outcome based on actual physiological responses, creating a closed-loop feedback mechanism for treatment prediction.

Inventive Principle:
Principle #23Feedback

2Reliability

If transcranial magnetic stimulation is applied to treat neuropsychiatric conditions, then therapeutic effect can be achieved, but treatment outcome cannot be predicted accurately

Engineering Contradiction:
Improvetreatment efficacyVSAvoidprediction system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system replaces complex clinical judgment and trial-and-error treatment approaches with automated computational analysis of brain electrical activity. By using algorithms to process EEG spectra, frequency band analysis, and brain network activity patterns, the system substitutes mechanical/clinical complexity with computational processing, achieving reliable treatment predictions without proportionally increasing physical device complexity.

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

Solution Approach 2:

The prediction system is designed to be universally applicable across different neuropsychiatric conditions (ADHD, depression, anxiety) and different TMS treatment protocols. By analyzing fundamental brain electrical activity patterns that are common across various conditions, the system achieves multi-functional prediction capability without requiring condition-specific or protocol-specific customization, reducing overall system complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Measurement precision

If multiple frequency bands and brain network activity patterns are analyzed, then treatment outcome prediction accuracy improves, but measurement and analysis complexity increases

Engineering Contradiction:
Improveprediction accuracyVSAvoidsignal analysis complexity
Core Design Contradiction:
Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The system segments the complex EEG signal into distinct frequency bands (delta, theta, alpha, beta, gamma) for separate analysis. By dividing the broad spectral range into discrete bands, the system can independently analyze characteristic patterns in each band without being overwhelmed by the full complexity of the raw signal, making the measurement and analysis process more manageable while maintaining high prediction accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system focuses analysis on specific, most-relevant frequency bands and brain network activity patterns that have been identified as particularly predictive of treatment outcome. Rather than analyzing every aspect of the EEG signal equally, the system applies partial action by concentrating computational resources on the most informative features, achieving high prediction accuracy without the excessive complexity of comprehensive signal analysis.

Inventive Principle:
Principle #16Partial or excessive 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 accurate prediction of treatment responses for neuropsychiatric conditions, such as ADHD, depression, and major depressive disorder, improving treatment efficacy and patient tolerance.

Implementation Method 1

The magnetic fields cause electric conduction in brain cells, and, as a consequence, generation of action potentials

Methodology Applied
Scientific EffectElectromagnetic induction: Electromagnetic Induction

Implementation Method 2

detect an electrophysiological signal of the subject, using the electrophysiological signal detecting electrodes

Methodology Applied
Scientific EffectElectrical signal detection: Electric Field

Data Source

PatentEP3532162B1Apparatus for predicting therapy outcome
Publication Date: 2024.07.24 BRAINSWAY
  • EP3532162B1 patent drawingFigure 1
  • EP3532162B1 patent drawingFigure 2
  • EP3532162B1 patent drawingFigure 3A~3C

AI summary

Apparatus and methods are described for use with electrophysiological signal detecting electrodes (14), and a transcranial magnetic stimulation device (10). A computer processor (16) drives the transcranial stimulation device to apply one or more pulses of transcranial magnetic stimulation to a subject. Within a given time period of applying one of the one or more pulses of transcranial magnetic stimulation to the subject, the computer processor detects an electrophysiological signal of the subject, using the electrophysiological signal detecting electrodes (14). At least partially in response thereto, the computer processor predicts an outcome of treating the subject for a neuropsychiatric condition, using a given therapy, and generates an output on an output device (18) in response to the predicted outcome. Other applications are also described.