Brain Network Mapping via MRI and ECoG Machine Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Traditional functional brain mapping techniques, such as direct electrocortical stimulation, are invasive and can cause adverse effects, and may miss critical brain areas, leading to discomfort and morbidity for patients, while also being time-consuming and potentially inaccurate in identifying eloquent brain functions.

Innovation Solution

A system and method using magnetic resonance imaging (MRI) data and electrocorticography (ECoG) recordings to generate network connectivity metrics and differentiate between critical and non-critical nodes in the brain network, predicting critical nodes for language and speech functions using machine learning classifiers based on network signatures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If direct electrical stimulation is used to map critical brain areas, then functional brain mapping can be performed, but adverse effects such as seizures and patient discomfort occur

Engineering Contradiction:

Improveidentification accuracy of critical brain areasVSAvoidadverse effects including seizures and patient discomfort

Core Design Contradiction:

Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent replaces the mechanical/electrical stimulation system with a computational analysis system. Instead of applying electrical stimuli to identify critical areas, the system uses machine learning algorithms to analyze neuroimaging data and predict critical brain regions, thereby eliminating seizures and patient discomfort while maintaining identification accuracy

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces machine learning models as an intermediary between neuroimaging data and critical area identification. The machine learning decoder processes imaging features to predict critical nodes, serving as a mediator that avoids direct electrical stimulation of brain tissue while still achieving functional mapping

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If direct electrical stimulation is used to identify eloquent functions, then critical nodes can be mapped, but the process is time-consuming

Engineering Contradiction:

Improveidentification accuracy of critical brain areasVSAvoidsurgical time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs critical area identification before surgery using preoperative neuroimaging data and machine learning analysis. By predicting critical nodes in advance, the system eliminates the need for time-consuming intraoperative stimulation mapping, thereby reducing surgical time while maintaining identification accuracy

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces the time-consuming electrical stimulation process with rapid computational analysis of preoperative imaging data, enabling critical area identification to be completed outside the operating room and significantly reducing surgical time

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If direct electrical stimulation is used to map brain functions, then critical areas can be identified, but some critical areas may be missed

Engineering Contradiction:

Improvecompleteness of critical area identificationVSAvoididentification accuracy of critical brain areas

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent uses machine learning models trained on multiple neuroimaging modalities (functional MRI, diffusion MRI, structural MRI) to capture diverse functional and structural information. This multi-functional approach allows the system to identify critical areas that might be missed by single-modality electrical stimulation, improving both completeness and accuracy

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent creates a computational model that replicates and extends the functionality of electrical stimulation mapping. The machine learning decoder processes multiple imaging dimensions simultaneously to predict critical nodes, providing a more comprehensive map than traditional stimulation methods while improving identification accuracy

Inventive Principle:

Principle #26Copying

Data Source

PatentUS20230123659A1Mapping critical brain sites using intracranial electrophysiology and machine learning

Publication Date: 2023.04.20 THE TRUSTEES OF INDIANA UNIV

AI summary

A system for performing functional brain mapping includes a memory configured to store first data from a magnetic resonance imaging (MRI) system and second data from electrodes. The system also includes a processor operatively coupled to the memory and configured to identify first edges in a brain network based on the first data from the MRI and second edges in the brain network based on the second data from the electrodes. The processor is configured to determine, based on the first edges and the second edges, connectivity metrics for the brain network. The processor is also configured to generate, based at least in part on the connectivity metrics, a decoder that differentiates between critical nodes and non-critical nodes in the brain network.