Brain Network Visualization via Parcellation Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current interfaces for neurosurgeons are limited in clinical assistance, as they display too many brain tracts, making it difficult to determine which tracts are connected and relevant to specific functions, leading to challenges in identifying conditions, planning surgeries, and understanding brain operations.

Innovation Solution

A method and system for generating a graphical representation of a brain network using MRI and diffusion tensor images, allowing users to select and visualize specific networks and tracts corresponding to particular functions or structures, by determining parcellations and generating surfaces representing these areas using three-dimensional coordinates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If DTI images display all brain tracts, then complete structural information is provided, but it becomes difficult to identify specific connected tracts and their functional relevance

Engineering Contradiction:
Improvebrain tract connection informationVSAvoididentification of specific tracts
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The patent segments the complex DTI brain tract data by creating parcellations that divide the brain into distinct functional regions. Each parcellation groups related tracts together, allowing neurosurgeons to selectively examine specific functional networks rather than being overwhelmed by all tracts simultaneously. This segmentation enables focused analysis of tract connections relevant to particular brain functions while maintaining access to complete structural information.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces parcellations as an intermediary layer between the raw DTI tract data and the neurosurgeon's analysis. These parcellations act as organizational mediators that structure the complex tract information into manageable functional groups, making it easier to identify and study specific tract connections without losing the underlying complete structural data.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of operation

If parcellations are used to organize brain regions, then tract identification becomes easier, but the system complexity increases

Engineering Contradiction:
Improvetract identificationVSAvoidsystem complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent implements a universal parcellation framework that can be applied across different DTI datasets and clinical scenarios. The parcellation system serves multiple functions: organizing tracts by functional region, enabling selective visualization, facilitating surgical planning, and supporting research analysis. This multi-functional approach justifies the system complexity by providing broad clinical utility across various neurological applications.

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

3Loss of information

If all brain tracts are visualized, then comprehensive structural data is available, but clinical decision-making becomes slower

Engineering Contradiction:
Improvestructural data completenessVSAvoidclinical decision time
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The patent implements dynamic visualization capabilities that allow neurosurgeons to interactively adjust the level of detail displayed. The system can dynamically switch between displaying all tracts, displaying only tracts within selected parcellations, or highlighting specific tract bundles relevant to the clinical question. This dynamic adaptation enables rapid clinical decision-making by presenting only the necessary information at each stage of analysis while maintaining access to complete structural data when needed.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS11704870B2Differential brain network analysis
Publication Date: 2023.07.18 OMNISCIENT NEUROTECH PTY LTD
  • US11704870B2 patent drawing
  • US11704870B2 patent drawing
  • US11704870B2 patent drawing

AI summary

A system and method of generating a graphical representation of a network of a subject human brain. The method comprises receiving, via a user interface, a selection of the network of the subject brain; determining, based on an MRI image of the subject brain and one or more identifiers associated with the selection, one or more parcellations of the subject brain (405); determining, using three-dimensional coordinates associated with each parcellation, corresponding tracts in a diffusion tensor image of the brain (425); and generating a graphical representation of the selected network (430), the graphical representation including at least one of (i) one or more surfaces representing the one or more parcellations, each surface generated using the coordinates, and (ii) the determined tracts.