-
Tytuł:
-
A multitask classification framework based on vision transformer for predicting molecular expressions of glioma.
-
Autorzy:
-
Xu Q; Department of Radiology, The Affiliated Hospital of Xuzhou Medical University, Xuzhou City, Jiangsu Province 221002, China.
Xu QQ; School of Medical Information and Engineering, Xuzhou Medical University, Xuzhou City, Jiangsu Province 221002, China.
Shi N; School of Medical Imaging, Xuzhou Medical University, Xuzhou City, Jiangsu Province 221002, China.
Dong LN; Department of Radiology, The Affiliated Hospital of Xuzhou Medical University, Xuzhou City, Jiangsu Province 221002, China.
Zhu H; School of Medical Information and Engineering, Xuzhou Medical University, Xuzhou City, Jiangsu Province 221002, China. Electronic address: .
Xu K; Department of Radiology, The Affiliated Hospital of Xuzhou Medical University, Xuzhou City, Jiangsu Province 221002, China. Electronic address: .
-
Źródło:
-
European journal of radiology [Eur J Radiol] 2022 Dec; Vol. 157, pp. 110560. Date of Electronic Publication: 2022 Oct 17.
-
Typ publikacji:
-
Journal Article
-
Język:
-
English
-
Imprint Name(s):
-
Publication: Limerick : Elsevier Science Ireland Ltd
Original Publication: Stuttgart ; New York : Thieme, [c1981-
-
MeSH Terms:
-
Brain Neoplasms*/diagnostic imaging
Glioma*/diagnostic imaging
Humans ; Isocitrate Dehydrogenase/genetics ; Ki-67 Antigen ; Tumor Suppressor Protein p53 ; Neoplasm Grading ; Mutation ; Magnetic Resonance Imaging/methods
-
Contributed Indexing:
-
Keywords: Convolutional neural network; Deep learning; Glioma; Magnetic resonance imaging; Molecular expression; Vision Transformer
-
Substance Nomenclature:
-
EC 1.1.1.41 (Isocitrate Dehydrogenase)
0 (Ki-67 Antigen)
0 (Tumor Suppressor Protein p53)
-
Entry Date(s):
-
Date Created: 20221103 Date Completed: 20221205 Latest Revision: 20221205
-
Update Code:
-
20240105
-
DOI:
-
10.1016/j.ejrad.2022.110560
-
PMID:
-
36327857
-
Purpose: The purpose of this study is to develop a Vision Transformer model with multitask classification framework that is appropriate for predicting four molecular expressions of glioma simultaneously based on MR imaging.
Materials and Methods: A total of 188 glioma (grades II-IV) patients with an immunohistochemical diagnosis of IDH, MGMT, Ki67 and P53 expression were enrolled in our study. A Vision Transformer (ViT) model, including three independent networks based on T2WI, T1CWI and T2 + T1CWI (T2-net, T1C-net and TU-net), was developed for the prediction of four glioma molecular expressions simultaneously. To evaluate the model performance, the accuracy rate, recall, precision, F1-score, and area under the receiver operating characteristic curve (AUC) were calculated.
Results: The proposed ViT model achieved high accuracy in predicting IDH, MGMT, Ki67 and P53 expression in gliomas. Among the three networks using the ViT model, TU-net achieved the best results with the highest values of accuracy (range, 0.937-0.969), precision (range, 0.949-0.972), recall (range, 0.873-0.991), F1-score (range, 0.910-0.981) and AUC (range, 0.976-0.984). Comparisons were also made between our ViT model and convolutional neural network (CNN)-based models, and the proposed ViT model outperformed the existing CNN-based models.
Conclusion: Vision Transformer is a reliable approach for the prediction of glioma molecular biomarkers and can be a viable alternative to CNNs.
Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
(Copyright © 2022 Elsevier B.V. All rights reserved.)
