Tumor peripheral surface auditory nerve identification method based on machine learning

The invention discloses a tumor peripheral surface auditory nerve identification method based on machine learning. In order to solve the problem that an anatomical structure cannot be identified by using an atlas due to deformation of an auditory nerve caused by tumor compression during MRI fiber re...

Full description

Saved in:
Bibliographic Details
Main Authors HUANG JIAHAO, CHEN SHENGWEI, QIU XIANG, WANG JIAFENG, YUAN SHAONAN, LU XINGZHOU, FENG YUANJING
Format Patent
LanguageChinese
English
Published 13.08.2021
Subjects
CALCULATING
COMPUTING
COUNTING
HANDLING RECORD CARRIERS
PHYSICS
PRESENTATION OF DATA
RECOGNITION OF DATA
RECORD CARRIERS
Online AccessGet full text

Cover

More Information
Summary:The invention discloses a tumor peripheral surface auditory nerve identification method based on machine learning. In order to solve the problem that an anatomical structure cannot be identified by using an atlas due to deformation of an auditory nerve caused by tumor compression during MRI fiber reconstruction of the patient, a method for only extracting a region of interest in order to eliminate interference of other fiber bundles in a data processing stage and converting a fiber bundle identification problem into a voxel classification problem is provided. Eight types of important features are extracted from each voxel, a feature sample set formed by the voxels is trained to obtain an optimal learning model, and the voxels of a test image are divided into three types of facial auditory nerves, auditory tumors and brainstems through the model; pseudo-color processing is carried out on a prediction result to restore the three-dimensional image to obtain a position relation among the three parts. 一种基于机器学习的瘤周面
Bibliography:Application Number: CN202110393849