Epilepsy recognition and operation curative effect prediction system based on graph neural network

The invention provides an epilepsy recognition and operation curative effect prediction system based on a graph neural network aiming at the limitation that a traditional graph neural network cannot fully utilize indirect connection and time dynamics of a brain function network, and the epilepsy rec...

Full description

Saved in:
Bibliographic Details
Main Authors WANG XUYANG, LI RONG, ZHANG JUNQUAN, LIU HAOXIANG
Format Patent
LanguageChinese
English
Published 12.07.2024
Subjects
CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
DIAGNOSIS
HUMAN NECESSITIES
HYGIENE
IDENTIFICATION
IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
MEDICAL OR VETERINARY SCIENCE
PHYSICS
SURGERY
Online AccessGet full text

Cover

More Information
Summary:The invention provides an epilepsy recognition and operation curative effect prediction system based on a graph neural network aiming at the limitation that a traditional graph neural network cannot fully utilize indirect connection and time dynamics of a brain function network, and the epilepsy recognition and operation curative effect prediction system is used for epilepsy brain function mode recognition and individual operation curative effect prediction. According to the method, feature vectors of nodes and edges are automatically learned by combining a graph attention mechanism of node and edge levels, information propagation among multiple nodes is promoted, and the spatial representation capability is improved. The result shows that the average classification accuracy of the model between the temporal lobe epilepsy and a normal control group is 85.52%, and the area of a subject under an operation characteristic curve is 0.913. In addition, the nidus area of the temporal lobe epilepsy can be recognized
Bibliography:Application Number: CN202410474823