Deep Learning-Based Acute Ischemic Stroke Lesion Segmentation Method on Multimodal MR Images Using a Few Fully Labeled Subjects

Acute ischemic stroke (AIS) has been a common threat to human health and may lead to severe outcomes without proper and prompt treatment. To precisely diagnose AIS, it is of paramount importance to quantitatively evaluate the AIS lesions. By adopting a convolutional neural network (CNN), many automa...

Full description

Bibliographic Details
Main Authors: Bin Zhao, Zhiyang Liu, Guohua Liu, Chen Cao, Song Jin, Hong Wu, Shuxue Ding
Format: Article
Language:English
Published: Hindawi Limited 2021-01-01
Series:Computational and Mathematical Methods in Medicine
Online Access:http://dx.doi.org/10.1155/2021/3628179
id doaj-68237dfdeb2a4e0aba48901af172cad8
record_format Article
spelling doaj-68237dfdeb2a4e0aba48901af172cad82021-02-15T12:52:57ZengHindawi LimitedComputational and Mathematical Methods in Medicine1748-670X1748-67182021-01-01202110.1155/2021/36281793628179Deep Learning-Based Acute Ischemic Stroke Lesion Segmentation Method on Multimodal MR Images Using a Few Fully Labeled SubjectsBin Zhao0Zhiyang Liu1Guohua Liu2Chen Cao3Song Jin4Hong Wu5Shuxue Ding6Tianjin Key Laboratory of Optoelectronic Sensor and Sensing Network Technology, College of Electronic Information and Optical Engineering, Nankai University, Tianjin 300350, ChinaTianjin Key Laboratory of Optoelectronic Sensor and Sensing Network Technology, College of Electronic Information and Optical Engineering, Nankai University, Tianjin 300350, ChinaTianjin Key Laboratory of Optoelectronic Sensor and Sensing Network Technology, College of Electronic Information and Optical Engineering, Nankai University, Tianjin 300350, ChinaKey Laboratory for Cerebral Artery and Neural Degeneration of Tianjin, Department of Medical Imaging, Tianjin Huanhu Hospital, Tianjin 300350, ChinaKey Laboratory for Cerebral Artery and Neural Degeneration of Tianjin, Department of Medical Imaging, Tianjin Huanhu Hospital, Tianjin 300350, ChinaTianjin Key Laboratory of Optoelectronic Sensor and Sensing Network Technology, College of Electronic Information and Optical Engineering, Nankai University, Tianjin 300350, ChinaTianjin Key Laboratory of Optoelectronic Sensor and Sensing Network Technology, College of Electronic Information and Optical Engineering, Nankai University, Tianjin 300350, ChinaAcute ischemic stroke (AIS) has been a common threat to human health and may lead to severe outcomes without proper and prompt treatment. To precisely diagnose AIS, it is of paramount importance to quantitatively evaluate the AIS lesions. By adopting a convolutional neural network (CNN), many automatic methods for ischemic stroke lesion segmentation on magnetic resonance imaging (MRI) have been proposed. However, most CNN-based methods should be trained on a large amount of fully labeled subjects, and the label annotation is a labor-intensive and time-consuming task. Therefore, in this paper, we propose to use a mixture of many weakly labeled and a few fully labeled subjects to relieve the thirst of fully labeled subjects. In particular, a multifeature map fusion network (MFMF-Network) with two branches is proposed, where hundreds of weakly labeled subjects are used to train the classification branch, and several fully labeled subjects are adopted to tune the segmentation branch. By training on 398 weakly labeled and 5 fully labeled subjects, the proposed method is able to achieve a mean dice coefficient of 0.699±0.128 on a test set with 179 subjects. The lesion-wise and subject-wise metrics are also evaluated, where a lesion-wise F1 score of 0.886 and a subject-wise detection rate of 1 are achieved.http://dx.doi.org/10.1155/2021/3628179
collection DOAJ
language English
format Article
sources DOAJ
author Bin Zhao
Zhiyang Liu
Guohua Liu
Chen Cao
Song Jin
Hong Wu
Shuxue Ding
spellingShingle Bin Zhao
Zhiyang Liu
Guohua Liu
Chen Cao
Song Jin
Hong Wu
Shuxue Ding
Deep Learning-Based Acute Ischemic Stroke Lesion Segmentation Method on Multimodal MR Images Using a Few Fully Labeled Subjects
Computational and Mathematical Methods in Medicine
author_facet Bin Zhao
Zhiyang Liu
Guohua Liu
Chen Cao
Song Jin
Hong Wu
Shuxue Ding
author_sort Bin Zhao
title Deep Learning-Based Acute Ischemic Stroke Lesion Segmentation Method on Multimodal MR Images Using a Few Fully Labeled Subjects
title_short Deep Learning-Based Acute Ischemic Stroke Lesion Segmentation Method on Multimodal MR Images Using a Few Fully Labeled Subjects
title_full Deep Learning-Based Acute Ischemic Stroke Lesion Segmentation Method on Multimodal MR Images Using a Few Fully Labeled Subjects
title_fullStr Deep Learning-Based Acute Ischemic Stroke Lesion Segmentation Method on Multimodal MR Images Using a Few Fully Labeled Subjects
title_full_unstemmed Deep Learning-Based Acute Ischemic Stroke Lesion Segmentation Method on Multimodal MR Images Using a Few Fully Labeled Subjects
title_sort deep learning-based acute ischemic stroke lesion segmentation method on multimodal mr images using a few fully labeled subjects
publisher Hindawi Limited
series Computational and Mathematical Methods in Medicine
issn 1748-670X
1748-6718
publishDate 2021-01-01
description Acute ischemic stroke (AIS) has been a common threat to human health and may lead to severe outcomes without proper and prompt treatment. To precisely diagnose AIS, it is of paramount importance to quantitatively evaluate the AIS lesions. By adopting a convolutional neural network (CNN), many automatic methods for ischemic stroke lesion segmentation on magnetic resonance imaging (MRI) have been proposed. However, most CNN-based methods should be trained on a large amount of fully labeled subjects, and the label annotation is a labor-intensive and time-consuming task. Therefore, in this paper, we propose to use a mixture of many weakly labeled and a few fully labeled subjects to relieve the thirst of fully labeled subjects. In particular, a multifeature map fusion network (MFMF-Network) with two branches is proposed, where hundreds of weakly labeled subjects are used to train the classification branch, and several fully labeled subjects are adopted to tune the segmentation branch. By training on 398 weakly labeled and 5 fully labeled subjects, the proposed method is able to achieve a mean dice coefficient of 0.699±0.128 on a test set with 179 subjects. The lesion-wise and subject-wise metrics are also evaluated, where a lesion-wise F1 score of 0.886 and a subject-wise detection rate of 1 are achieved.
url http://dx.doi.org/10.1155/2021/3628179
work_keys_str_mv AT binzhao deeplearningbasedacuteischemicstrokelesionsegmentationmethodonmultimodalmrimagesusingafewfullylabeledsubjects
AT zhiyangliu deeplearningbasedacuteischemicstrokelesionsegmentationmethodonmultimodalmrimagesusingafewfullylabeledsubjects
AT guohualiu deeplearningbasedacuteischemicstrokelesionsegmentationmethodonmultimodalmrimagesusingafewfullylabeledsubjects
AT chencao deeplearningbasedacuteischemicstrokelesionsegmentationmethodonmultimodalmrimagesusingafewfullylabeledsubjects
AT songjin deeplearningbasedacuteischemicstrokelesionsegmentationmethodonmultimodalmrimagesusingafewfullylabeledsubjects
AT hongwu deeplearningbasedacuteischemicstrokelesionsegmentationmethodonmultimodalmrimagesusingafewfullylabeledsubjects
AT shuxueding deeplearningbasedacuteischemicstrokelesionsegmentationmethodonmultimodalmrimagesusingafewfullylabeledsubjects
_version_ 1714866874121977856