Overall Survival Prediction Using Conventional MRI Features

Gliomas are common primary brain malignancies. The sub-regions of gliomas are depicted by MRI scans, reflecting varying biological properties. These properties have effect on the diagnosis of neurosurgeons on whether or what kind of resection should be done. The survival days after gross total resec...

Full description

Saved in:
Bibliographic Details
Published inBrainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries pp. 244 - 254
Main Authors Ren, Yanhao, Sun, Pin, Lu, Wenlian
Format Book Chapter
LanguageEnglish
Published Cham Springer International Publishing
SeriesLecture Notes in Computer Science
Subjects
Conventional MRI features
MRI scans
Semi-auto segmentation
Support vector regression
Survival prediction
Online AccessGet full text

Cover

More Information
Summary:Gliomas are common primary brain malignancies. The sub-regions of gliomas are depicted by MRI scans, reflecting varying biological properties. These properties have effect on the diagnosis of neurosurgeons on whether or what kind of resection should be done. The survival days after gross total resection is also of great concern. In this paper, we propose a semi-auto method for segmentation, and extract features from slices of MRI scans, including conventional MRI features and clinical features. 13 features of a subject are selected finally and a support vector regression is used to fit with the training data.
ISBN:3030466426
9783030466428
ISSN:0302-9743
1611-3349
DOI:10.1007/978-3-030-46643-5_24