Cargando…

A Deep Multi-Task Learning Framework for Brain Tumor Segmentation

Glioma is the most common primary central nervous system tumor, accounting for about half of all intracranial primary tumors. As a non-invasive examination method, MRI has an extremely important guiding role in the clinical intervention of tumors. However, manually segmenting brain tumors from MRI r...

Descripción completa

Detalles Bibliográficos
Autores principales: Huang, He, Yang, Guang, Zhang, Wenbo, Xu, Xiaomei, Yang, Weiji, Jiang, Weiwei, Lai, Xiaobo
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8212784/
https://www.ncbi.nlm.nih.gov/pubmed/34150660
http://dx.doi.org/10.3389/fonc.2021.690244
_version_ 1783709705580838912
author Huang, He
Yang, Guang
Zhang, Wenbo
Xu, Xiaomei
Yang, Weiji
Jiang, Weiwei
Lai, Xiaobo
author_facet Huang, He
Yang, Guang
Zhang, Wenbo
Xu, Xiaomei
Yang, Weiji
Jiang, Weiwei
Lai, Xiaobo
author_sort Huang, He
collection PubMed
description Glioma is the most common primary central nervous system tumor, accounting for about half of all intracranial primary tumors. As a non-invasive examination method, MRI has an extremely important guiding role in the clinical intervention of tumors. However, manually segmenting brain tumors from MRI requires a lot of time and energy for doctors, which affects the implementation of follow-up diagnosis and treatment plans. With the development of deep learning, medical image segmentation is gradually automated. However, brain tumors are easily confused with strokes and serious imbalances between classes make brain tumor segmentation one of the most difficult tasks in MRI segmentation. In order to solve these problems, we propose a deep multi-task learning framework and integrate a multi-depth fusion module in the framework to accurately segment brain tumors. In this framework, we have added a distance transform decoder based on the V-Net, which can make the segmentation contour generated by the mask decoder more accurate and reduce the generation of rough boundaries. In order to combine the different tasks of the two decoders, we weighted and added their corresponding loss functions, where the distance map prediction regularized the mask prediction. At the same time, the multi-depth fusion module in the encoder can enhance the ability of the network to extract features. The accuracy of the model will be evaluated online using the multispectral MRI records of the BraTS 2018, BraTS 2019, and BraTS 2020 datasets. This method obtains high-quality segmentation results, and the average Dice is as high as 78%. The experimental results show that this model has great potential in segmenting brain tumors automatically and accurately.
format Online
Article
Text
id pubmed-8212784
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-82127842021-06-19 A Deep Multi-Task Learning Framework for Brain Tumor Segmentation Huang, He Yang, Guang Zhang, Wenbo Xu, Xiaomei Yang, Weiji Jiang, Weiwei Lai, Xiaobo Front Oncol Oncology Glioma is the most common primary central nervous system tumor, accounting for about half of all intracranial primary tumors. As a non-invasive examination method, MRI has an extremely important guiding role in the clinical intervention of tumors. However, manually segmenting brain tumors from MRI requires a lot of time and energy for doctors, which affects the implementation of follow-up diagnosis and treatment plans. With the development of deep learning, medical image segmentation is gradually automated. However, brain tumors are easily confused with strokes and serious imbalances between classes make brain tumor segmentation one of the most difficult tasks in MRI segmentation. In order to solve these problems, we propose a deep multi-task learning framework and integrate a multi-depth fusion module in the framework to accurately segment brain tumors. In this framework, we have added a distance transform decoder based on the V-Net, which can make the segmentation contour generated by the mask decoder more accurate and reduce the generation of rough boundaries. In order to combine the different tasks of the two decoders, we weighted and added their corresponding loss functions, where the distance map prediction regularized the mask prediction. At the same time, the multi-depth fusion module in the encoder can enhance the ability of the network to extract features. The accuracy of the model will be evaluated online using the multispectral MRI records of the BraTS 2018, BraTS 2019, and BraTS 2020 datasets. This method obtains high-quality segmentation results, and the average Dice is as high as 78%. The experimental results show that this model has great potential in segmenting brain tumors automatically and accurately. Frontiers Media S.A. 2021-06-04 /pmc/articles/PMC8212784/ /pubmed/34150660 http://dx.doi.org/10.3389/fonc.2021.690244 Text en Copyright © 2021 Huang, Yang, Zhang, Xu, Yang, Jiang and Lai https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Oncology
Huang, He
Yang, Guang
Zhang, Wenbo
Xu, Xiaomei
Yang, Weiji
Jiang, Weiwei
Lai, Xiaobo
A Deep Multi-Task Learning Framework for Brain Tumor Segmentation
title A Deep Multi-Task Learning Framework for Brain Tumor Segmentation
title_full A Deep Multi-Task Learning Framework for Brain Tumor Segmentation
title_fullStr A Deep Multi-Task Learning Framework for Brain Tumor Segmentation
title_full_unstemmed A Deep Multi-Task Learning Framework for Brain Tumor Segmentation
title_short A Deep Multi-Task Learning Framework for Brain Tumor Segmentation
title_sort deep multi-task learning framework for brain tumor segmentation
topic Oncology
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8212784/
https://www.ncbi.nlm.nih.gov/pubmed/34150660
http://dx.doi.org/10.3389/fonc.2021.690244
work_keys_str_mv AT huanghe adeepmultitasklearningframeworkforbraintumorsegmentation
AT yangguang adeepmultitasklearningframeworkforbraintumorsegmentation
AT zhangwenbo adeepmultitasklearningframeworkforbraintumorsegmentation
AT xuxiaomei adeepmultitasklearningframeworkforbraintumorsegmentation
AT yangweiji adeepmultitasklearningframeworkforbraintumorsegmentation
AT jiangweiwei adeepmultitasklearningframeworkforbraintumorsegmentation
AT laixiaobo adeepmultitasklearningframeworkforbraintumorsegmentation
AT huanghe deepmultitasklearningframeworkforbraintumorsegmentation
AT yangguang deepmultitasklearningframeworkforbraintumorsegmentation
AT zhangwenbo deepmultitasklearningframeworkforbraintumorsegmentation
AT xuxiaomei deepmultitasklearningframeworkforbraintumorsegmentation
AT yangweiji deepmultitasklearningframeworkforbraintumorsegmentation
AT jiangweiwei deepmultitasklearningframeworkforbraintumorsegmentation
AT laixiaobo deepmultitasklearningframeworkforbraintumorsegmentation