Improving the diagnosis of acute ischemic stroke on non-contrast CT using deep learning: a multicenter study

Key points The model simultaneously segments infarcts and estimates ASPECTS by using baseline CT. A mirror-assembly module plus dual-path DCNN model improved the segmentation efficiency. We evaluated the model in a multi-reader and multicenter (MRMC) setting.

Bibliographic Details
Main Authors: Weidao Chen, Jiangfen Wu, Ren Wei, Shuang Wu, Chen Xia, Dawei Wang, Daliang Liu, Longmei Zheng, Tianyu Zou, Ruijiang Li, Xianrong Qi, Xiaotong Zhang
Format: Article
Language:English
Published: SpringerOpen 2022-12-01
Series:Insights into Imaging
Subjects:
Online Access:https://doi.org/10.1186/s13244-022-01331-3
_version_ 1811203559554809856
author Weidao Chen
Jiangfen Wu
Ren Wei
Shuang Wu
Chen Xia
Dawei Wang
Daliang Liu
Longmei Zheng
Tianyu Zou
Ruijiang Li
Xianrong Qi
Xiaotong Zhang
author_facet Weidao Chen
Jiangfen Wu
Ren Wei
Shuang Wu
Chen Xia
Dawei Wang
Daliang Liu
Longmei Zheng
Tianyu Zou
Ruijiang Li
Xianrong Qi
Xiaotong Zhang
author_sort Weidao Chen
collection DOAJ
description Key points The model simultaneously segments infarcts and estimates ASPECTS by using baseline CT. A mirror-assembly module plus dual-path DCNN model improved the segmentation efficiency. We evaluated the model in a multi-reader and multicenter (MRMC) setting.
first_indexed 2024-04-12T02:57:26Z
format Article
id doaj.art-a269ba8efc054989988712c7aa286b8c
institution Directory Open Access Journal
issn 1869-4101
language English
last_indexed 2024-04-12T02:57:26Z
publishDate 2022-12-01
publisher SpringerOpen
record_format Article
series Insights into Imaging
spelling doaj.art-a269ba8efc054989988712c7aa286b8c2022-12-22T03:50:46ZengSpringerOpenInsights into Imaging1869-41012022-12-0113111210.1186/s13244-022-01331-3Improving the diagnosis of acute ischemic stroke on non-contrast CT using deep learning: a multicenter studyWeidao Chen0Jiangfen Wu1Ren Wei2Shuang Wu3Chen Xia4Dawei Wang5Daliang Liu6Longmei Zheng7Tianyu Zou8Ruijiang Li9Xianrong Qi10Xiaotong Zhang11Interdisciplinary Institute of Neuroscience and Technology, College of Biomedical Engineering and Instrument Science, Zhejiang UniversityAcademy for Advanced Interdisciplinary Studies, Peking UniversityInfervision Institute of ResearchInfervision Institute of ResearchInfervision Institute of ResearchInfervision Institute of ResearchLiaocheng People’s HospitalMedical Imaging Center, Ankang Central HospitalWeihai Municipal HospitalDepartment of Radiation Oncology, Stanford University School of MedicineSchool of Pharmaceutical Sciences, Peking UniversityInterdisciplinary Institute of Neuroscience and Technology, College of Biomedical Engineering and Instrument Science, Zhejiang UniversityKey points The model simultaneously segments infarcts and estimates ASPECTS by using baseline CT. A mirror-assembly module plus dual-path DCNN model improved the segmentation efficiency. We evaluated the model in a multi-reader and multicenter (MRMC) setting.https://doi.org/10.1186/s13244-022-01331-3Deep learningNon-contrast computed tomographyEarly ischemic changesAlberta Stroke Program Early Computed Tomography ScoreMulti-reader and multi-center studyAcute ischemic stroke
spellingShingle Weidao Chen
Jiangfen Wu
Ren Wei
Shuang Wu
Chen Xia
Dawei Wang
Daliang Liu
Longmei Zheng
Tianyu Zou
Ruijiang Li
Xianrong Qi
Xiaotong Zhang
Improving the diagnosis of acute ischemic stroke on non-contrast CT using deep learning: a multicenter study
Insights into Imaging
Deep learning
Non-contrast computed tomography
Early ischemic changes
Alberta Stroke Program Early Computed Tomography Score
Multi-reader and multi-center study
Acute ischemic stroke
title Improving the diagnosis of acute ischemic stroke on non-contrast CT using deep learning: a multicenter study
title_full Improving the diagnosis of acute ischemic stroke on non-contrast CT using deep learning: a multicenter study
title_fullStr Improving the diagnosis of acute ischemic stroke on non-contrast CT using deep learning: a multicenter study
title_full_unstemmed Improving the diagnosis of acute ischemic stroke on non-contrast CT using deep learning: a multicenter study
title_short Improving the diagnosis of acute ischemic stroke on non-contrast CT using deep learning: a multicenter study
title_sort improving the diagnosis of acute ischemic stroke on non contrast ct using deep learning a multicenter study
topic Deep learning
Non-contrast computed tomography
Early ischemic changes
Alberta Stroke Program Early Computed Tomography Score
Multi-reader and multi-center study
Acute ischemic stroke
url https://doi.org/10.1186/s13244-022-01331-3
work_keys_str_mv AT weidaochen improvingthediagnosisofacuteischemicstrokeonnoncontrastctusingdeeplearningamulticenterstudy
AT jiangfenwu improvingthediagnosisofacuteischemicstrokeonnoncontrastctusingdeeplearningamulticenterstudy
AT renwei improvingthediagnosisofacuteischemicstrokeonnoncontrastctusingdeeplearningamulticenterstudy
AT shuangwu improvingthediagnosisofacuteischemicstrokeonnoncontrastctusingdeeplearningamulticenterstudy
AT chenxia improvingthediagnosisofacuteischemicstrokeonnoncontrastctusingdeeplearningamulticenterstudy
AT daweiwang improvingthediagnosisofacuteischemicstrokeonnoncontrastctusingdeeplearningamulticenterstudy
AT daliangliu improvingthediagnosisofacuteischemicstrokeonnoncontrastctusingdeeplearningamulticenterstudy
AT longmeizheng improvingthediagnosisofacuteischemicstrokeonnoncontrastctusingdeeplearningamulticenterstudy
AT tianyuzou improvingthediagnosisofacuteischemicstrokeonnoncontrastctusingdeeplearningamulticenterstudy
AT ruijiangli improvingthediagnosisofacuteischemicstrokeonnoncontrastctusingdeeplearningamulticenterstudy
AT xianrongqi improvingthediagnosisofacuteischemicstrokeonnoncontrastctusingdeeplearningamulticenterstudy
AT xiaotongzhang improvingthediagnosisofacuteischemicstrokeonnoncontrastctusingdeeplearningamulticenterstudy