Inter-vendor performance of deep learning in segmenting acute ischemic lesions on diffusion-weighted imaging: a multicenter study

Abstract There is little evidence on the applicability of deep learning (DL) in the segmentation of acute ischemic lesions on diffusion-weighted imaging (DWI) between magnetic resonance imaging (MRI) scanners of different manufacturers. We retrospectively included DWI data of patients with acute isc...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Deniz Alis, Mert Yergin, Ceren Alis, Cagdas Topel, Ozan Asmakutlu, Omer Bagcilar, Yeseren Deniz Senli, Ahmet Ustundag, Vefa Salt, Sebahat Nacar Dogan, Murat Velioglu, Hakan Hatem Selcuk, Batuhan Kara, Ilkay Oksuz, Osman Kizilkilic, Ercan Karaarslan
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/897683b2837b4094b80466289ed7a30d
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:897683b2837b4094b80466289ed7a30d
record_format dspace
spelling oai:doaj.org-article:897683b2837b4094b80466289ed7a30d2021-12-02T17:40:49ZInter-vendor performance of deep learning in segmenting acute ischemic lesions on diffusion-weighted imaging: a multicenter study10.1038/s41598-021-91467-x2045-2322https://doaj.org/article/897683b2837b4094b80466289ed7a30d2021-06-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-91467-xhttps://doaj.org/toc/2045-2322Abstract There is little evidence on the applicability of deep learning (DL) in the segmentation of acute ischemic lesions on diffusion-weighted imaging (DWI) between magnetic resonance imaging (MRI) scanners of different manufacturers. We retrospectively included DWI data of patients with acute ischemic lesions from six centers. Dataset A (n = 2986) and B (n = 3951) included data from Siemens and GE MRI scanners, respectively. The datasets were split into the training (80%), validation (10%), and internal test (10%) sets, and six neuroradiologists created ground-truth masks. Models A and B were the proposed neural networks trained on datasets A and B. The models subsequently fine-tuned across the datasets using their validation data. Another radiologist performed the segmentation on the test sets for comparisons. The median Dice scores of models A and B were 0.858 and 0.857 for the internal tests, which were non-inferior to the radiologist’s performance, but demonstrated lower performance than the radiologist on the external tests. Fine-tuned models A and B achieved median Dice scores of 0.832 and 0.846, which were non-inferior to the radiologist's performance on the external tests. The present work shows that the inter-vendor operability of deep learning for the segmentation of ischemic lesions on DWI might be enhanced via transfer learning; thereby, their clinical applicability and generalizability could be improved.Deniz AlisMert YerginCeren AlisCagdas TopelOzan AsmakutluOmer BagcilarYeseren Deniz SenliAhmet UstundagVefa SaltSebahat Nacar DoganMurat VeliogluHakan Hatem SelcukBatuhan KaraIlkay OksuzOsman KizilkilicErcan KaraarslanNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-10 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Deniz Alis
Mert Yergin
Ceren Alis
Cagdas Topel
Ozan Asmakutlu
Omer Bagcilar
Yeseren Deniz Senli
Ahmet Ustundag
Vefa Salt
Sebahat Nacar Dogan
Murat Velioglu
Hakan Hatem Selcuk
Batuhan Kara
Ilkay Oksuz
Osman Kizilkilic
Ercan Karaarslan
Inter-vendor performance of deep learning in segmenting acute ischemic lesions on diffusion-weighted imaging: a multicenter study
description Abstract There is little evidence on the applicability of deep learning (DL) in the segmentation of acute ischemic lesions on diffusion-weighted imaging (DWI) between magnetic resonance imaging (MRI) scanners of different manufacturers. We retrospectively included DWI data of patients with acute ischemic lesions from six centers. Dataset A (n = 2986) and B (n = 3951) included data from Siemens and GE MRI scanners, respectively. The datasets were split into the training (80%), validation (10%), and internal test (10%) sets, and six neuroradiologists created ground-truth masks. Models A and B were the proposed neural networks trained on datasets A and B. The models subsequently fine-tuned across the datasets using their validation data. Another radiologist performed the segmentation on the test sets for comparisons. The median Dice scores of models A and B were 0.858 and 0.857 for the internal tests, which were non-inferior to the radiologist’s performance, but demonstrated lower performance than the radiologist on the external tests. Fine-tuned models A and B achieved median Dice scores of 0.832 and 0.846, which were non-inferior to the radiologist's performance on the external tests. The present work shows that the inter-vendor operability of deep learning for the segmentation of ischemic lesions on DWI might be enhanced via transfer learning; thereby, their clinical applicability and generalizability could be improved.
format article
author Deniz Alis
Mert Yergin
Ceren Alis
Cagdas Topel
Ozan Asmakutlu
Omer Bagcilar
Yeseren Deniz Senli
Ahmet Ustundag
Vefa Salt
Sebahat Nacar Dogan
Murat Velioglu
Hakan Hatem Selcuk
Batuhan Kara
Ilkay Oksuz
Osman Kizilkilic
Ercan Karaarslan
author_facet Deniz Alis
Mert Yergin
Ceren Alis
Cagdas Topel
Ozan Asmakutlu
Omer Bagcilar
Yeseren Deniz Senli
Ahmet Ustundag
Vefa Salt
Sebahat Nacar Dogan
Murat Velioglu
Hakan Hatem Selcuk
Batuhan Kara
Ilkay Oksuz
Osman Kizilkilic
Ercan Karaarslan
author_sort Deniz Alis
title Inter-vendor performance of deep learning in segmenting acute ischemic lesions on diffusion-weighted imaging: a multicenter study
title_short Inter-vendor performance of deep learning in segmenting acute ischemic lesions on diffusion-weighted imaging: a multicenter study
title_full Inter-vendor performance of deep learning in segmenting acute ischemic lesions on diffusion-weighted imaging: a multicenter study
title_fullStr Inter-vendor performance of deep learning in segmenting acute ischemic lesions on diffusion-weighted imaging: a multicenter study
title_full_unstemmed Inter-vendor performance of deep learning in segmenting acute ischemic lesions on diffusion-weighted imaging: a multicenter study
title_sort inter-vendor performance of deep learning in segmenting acute ischemic lesions on diffusion-weighted imaging: a multicenter study
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/897683b2837b4094b80466289ed7a30d
work_keys_str_mv AT denizalis intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT mertyergin intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT cerenalis intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT cagdastopel intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT ozanasmakutlu intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT omerbagcilar intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT yeserendenizsenli intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT ahmetustundag intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT vefasalt intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT sebahatnacardogan intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT muratvelioglu intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT hakanhatemselcuk intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT batuhankara intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT ilkayoksuz intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT osmankizilkilic intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
AT ercankaraarslan intervendorperformanceofdeeplearninginsegmentingacuteischemiclesionsondiffusionweightedimagingamulticenterstudy
_version_ 1718379716509958144