546 251

Cited 0 times in

Fully automated hybrid approach to predict the IDH mutation status of gliomas via deep learning and radiomics

 Yoon Seong Choi  ;  Sohi Bae  ;  Jong Hee Chang  ;  Seok-Gu Kang  ;  Se Hoon Kim  ;  Jinna Kim  ;  Tyler Hyungtaek Rim  ;  Seung Hong Choi  ;  Rajan Jain  ;  Seung-Koo Lee 
 NEURO-ONCOLOGY, Vol.23(2) : 304-313, 2021-02 
Journal Title
Issue Date
Brain Neoplasms* / diagnostic imaging ; Brain Neoplasms* / genetics ; Deep Learning* ; Glioma* / diagnostic imaging ; Glioma* / genetics ; Humans ; Isocitrate Dehydrogenase / genetics ; Magnetic Resonance Imaging ; Mutation ; Retrospective Studies
convolutional neural network ; glioma ; isocitrate dehydrogenase mutation ; magnetic resonance imaging ; radiomics
Background: Glioma prognosis depends on isocitrate dehydrogenase (IDH) mutation status. We aimed to predict the IDH status of gliomas from preoperative MR images using a fully automated hybrid approach with convolutional neural networks (CNNs) and radiomics.

Methods: We reviewed 1166 preoperative MR images of gliomas (grades II-IV) from Severance Hospital (n = 856), Seoul National University Hospital (SNUH; n = 107), and The Cancer Imaging Archive (TCIA; n = 203). The Severance set was subdivided into the development (n = 727) and internal test (n = 129) sets. Based on T1 postcontrast, T2, and fluid-attenuated inversion recovery images, a fully automated model was developed that comprised a CNN for tumor segmentation (Model 1) and CNN-based classifier for IDH status prediction (Model 2) that uses a hybrid approach based on 2D tumor images and radiomic features from 3D tumor shape and loci guided by Model 1. The trained model was tested on internal (a subset of the Severance set) and external (SNUH and TCIA) test sets.

Results: The CNN for tumor segmentation (Model 1) achieved a dice coefficient of 0.86-0.92 across datasets. Our hybrid model achieved accuracies of 93.8%, 87.9%, and 78.8%, with areas under the receiver operating characteristic curves of 0.96, 0.94, and 0.86 and areas under the precision-recall curves of 0.88, 0.82, and 0.81 in the internal test, SNUH, and TCIA sets, respectively.

Conclusions: Our fully automated hybrid model demonstrated the potential to be a highly reproducible and generalizable tool across different datasets for the noninvasive prediction of the IDH status of gliomas.
Files in This Item:
T202102294.pdf Download
Appears in Collections:
1. College of Medicine (의과대학) > Dept. of Neurosurgery (신경외과학교실) > 1. Journal Papers
1. College of Medicine (의과대학) > Dept. of Pathology (병리학교실) > 1. Journal Papers
1. College of Medicine (의과대학) > Dept. of Radiology (영상의학교실) > 1. Journal Papers
Yonsei Authors
Kang, Seok Gu(강석구) ORCID logo https://orcid.org/0000-0001-5676-2037
Kim, Se Hoon(김세훈) ORCID logo https://orcid.org/0000-0001-7516-7372
Kim, Jinna(김진아) ORCID logo https://orcid.org/0000-0002-9978-4356
Lee, Seung Koo(이승구) ORCID logo https://orcid.org/0000-0001-5646-4072
Chang, Jong Hee(장종희) ORCID logo https://orcid.org/0000-0003-1509-9800
Choi, Yoon Seong(최윤성)
사서에게 알리기


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.