Method and system for extracting multi-dimensional disconnection network region of symptom mapping

Patent metadata

Field	Value
Publication number	US-2024046454-A1
Application number	US-202318163425-A
Country	US
Kind code	A1
Filing date	Feb 2, 2023
Priority date	Aug 2, 2022
Publication date	Feb 8, 2024
Grant date	—

Abstract

A method and system for extracting a multi-dimensional disconnection network region of symptom mapping: registering a lesion image to a brain standard space; obtaining diffusion magnetic resonance images and resting-state functional magnetic resonance images of healthy control groups; constructing a structural disconnection weighting network corresponding to lesions using a fiber tracking method according to the lesion image in the brain standard space and the diffusion magnetic resonance images; constructing a functional significant disconnection network corresponding to the lesions using a cross-correlation verification method according to the lesion image in the brain standard space and the resting-state functional magnetic resonance images; and determining the multi-dimensional disconnection network region of the lesions of symptom mapping according to the structural disconnection weighting network and the functional significant disconnection network, where the multi-dimensional disconnection network region of the lesions is configured to locate network mapping of a brain lesion in the brain.

First claim

What is claimed is:

1 . A method for extracting a multi-dimensional disconnection network region of symptom mapping, comprising:
obtaining a lesion image, and registering the lesion image to a brain standard space; using brain images of healthy people with multiple ages and a balanced sex ratio as healthy control groups, and obtaining diffusion magnetic resonance images and resting-state functional magnetic resonance images of the healthy control groups; constructing a structural disconnection weighting network corresponding to lesions using a fiber tracking method according to the lesion image in the brain standard space and the diffusion magnetic resonance images, wherein the fiber tracking method comprises deterministic fiber tracking and probabilistic fiber tracking; constructing a functional significant disconnection network corresponding to the lesions using a cross-correlation verification method according to the lesion image in the brain standard space and the resting-state functional magnetic resonance images; and determining the multi-dimensional disconnection network region of the lesions of symptom mapping according to the structural disconnection weighting network and the functional significant disconnection network, wherein the multi-dimensional disconnection network region of the lesions is configured to locate network mapping of a brain lesion in the brain.

2 . The method for extracting a multi-dimensional disconnection network region of symptom mapping according to claim 1 , wherein a process of registering the lesion image to a brain standard space specifically comprises:
registering the lesion image to the brain standard space through linear transformation.

3 . The method for extracting a multi-dimensional disconnection network region of symptom mapping according to claim 1 , wherein a process of constructing a structural disconnection weighting network corresponding to lesions using a fiber tracking method according to the lesion image in the brain standard space and the diffusion magnetic resonance images specifically comprises:
performing pre-processing and diffusion weighted imaging modeling on the diffusion magnetic resonance images, and performing fiber tracking using the fiber tracking method to construct deterministic and probabilistic fiber tracking structural connection images in individual spaces of the healthy control groups; registering the lesion image in the brain standard space to an individual space of the healthy control group through linear transformation for any of the healthy control groups, and extracting a structural connection image of the lesions according to the deterministic and probabilistic fiber tracking structural connection images; calculating a weighted average of the structural connection image of the lesions to determine a structural disconnection weighting network in the individual spaces; registering the structural disconnection weighting network in the individual spaces to the brain standard space through linear transformation to determine a structural disconnection weighting network of the lesions corresponding to the healthy control groups; and constructing the structural disconnection weighting network corresponding to the lesions according to the structural disconnection weighting network of the lesions corresponding to all of the healthy control groups.

4 . The method for extracting a multi-dimensional disconnection network region of symptom mapping according to claim 1 , wherein a process of constructing a functional significant disconnection network corresponding to the lesions using a cross-correlation verification method according to the lesion image in the brain standard space and the resting-state functional magnetic resonance images specifically comprises:
pre-processing the resting-state functional magnetic resonance images to determine a brain functional signal image in the brain standard space; extracting an average signal of a range of the lesions by taking the lesion image as a region of interest (ROI) in the brain standard space; performing Pearson correlation between the average signal and signals of the rest of the whole brain based on the brain functional image to determine a functional connection value between the whole brain and a lesion region; calculating cross-correlation and out-of-order correlation values between the average signal and the signals of the rest of the whole brain using the cross-correlation verification method; retaining a cross-correlation value more than 100 times the out-of-order correlation value as a first cross-correlation value, and taking a functional connection value of a position corresponding to the first cross-correlation value as a functional significant disconnection network of the lesions corresponding to the healthy control groups; and constructing the functional significant disconnection network corresponding to the lesions according to the functional significant disconnection network of the lesions corresponding to all of the healthy control groups.

5 . A system for extracting a multi-dimensional disconnection network region of symptom mapping, comprising:
a registration module configured to obtain a lesion image, and register the lesion image to a brain standard space; an image obtaining module configured to use brain images of healthy people with multiple ages and a balanced sex ratio as healthy control groups, and obtain diffusion magnetic resonance images and resting-state functional magnetic resonance images of the healthy control groups; a structural disconnection weighting network construction module corresponding to lesions configured to construct a structural disconnection weighting network corresponding to lesions using a fiber tracking method according to the lesion image in the brain standard space and the diffusion magnetic resonance images, wherein the fiber tracking method comprises deterministic fiber tracking and probabilistic fiber tracking; a functional significant disconnection network construction module corresponding to lesions configured to construct a functional significant disconnection network corresponding to the lesions using a cross-correlation verification method according to the lesion image in the brain standard space and the resting-state functional magnetic resonance images; and a multi-dimensional disconnection network region determination module of lesions configured to determine the multi-dimensional disconnection network region of the lesions of symptom mapping according to the structural disconnection weighting network and the functional significant disconnection network, wherein the multi-dimensional disconnection network region of the lesions is configured to locate network mapping of a brain lesion in the brain.

6 . The system for extracting a multi-dimensional disconnection network region of symptom mapping according to claim 5 , wherein the registration module specifically comprises:
a registration unit configured to register the lesion image to the brain standard space through linear transformation.

7 . The system for extracting a multi-dimensional disconnection network region of symptom mapping according to claim 5 , wherein the structural disconnection weighting network construction module corresponding to lesions specifically comprises:
a deterministic and probabilistic fiber tracking structural connection image construction unit configured to perform pre-processing and diffusion weighted imaging modeling on the diffusion magnetic resonance images, and perform fiber tracking using the fiber tracking method to construct deterministic and probabilistic fiber tracking structural connection images in individual spaces of the healthy control groups; a structural connection image extract

Assignees

• Beijing Tiantan Hospital Capital Medical Univ
• Univ Beihang

Inventors

• Li Zixiao
• Liu Tao
• Liu Hao
• Ding Lingling
• Wang Yongjun

Classifications

• A61B5/7425

  Displaying combinations of multiple images regardless of image source, e.g. displaying a reference anatomical image with a live image · CPC title

• A61B5/4064

  Evaluating the brain (for intracranial pressure A61B5/031; for cerebral blood gases A61B5/14553; using EEG A61B5/369) · CPC title

• G06T7/0012 Primary

  Biomedical image inspection · CPC title

• A61B5/0042 Primary

  for the brain · CPC title

• G06T7/73

  using feature-based methods · CPC title