基于视觉Transformer的多任务分类框架预测胶质瘤分子表达
本研究纳入188例胶质瘤患者,开发基于T2WI、T1CWI及T2+T1CWI的视觉Transformer多任务模型,同时预测IDH、MGMT、Ki67和P53表达。TU-net表现最佳,准确率0.937-0.969,AUC 0.976-0.984,优于CNN模型。
要点
- ViT多任务模型可同时预测胶质瘤四个分子标志物,TU-net性能最优。
- TU-net的AUC达0.976-0.984,准确率0.937-0.969。
- ViT模型预测性能优于现有CNN模型。
核心结论
- 视觉Transformer是预测胶质瘤分子生物标志物的可靠方法,可作为CNN的可行替代方案。
研究设计与样本
- 回顾性研究纳入188例经免疫组化诊断IDH、MGMT、Ki67和P53表达的胶质瘤(II-IV级)患者。
- 构建基于T2WI、T1CWI及T2+T1CWI的三个独立ViT网络(T2-net、T1C-net、TU-net),同时预测四种分子表达。
诊断标准与关键数值
- TU-net预测四类分子表达的准确率范围0.937-0.969,精确率0.949-0.972,召回率0.873-0.991,F1分数0.910-0.981,AUC 0.976-0.984。
临床应用与阅片要点
- 可利用ViT多任务模型基于常规MRI序列同时预测胶质瘤IDH、MGMT、Ki67和P53表达,辅助术前分子分型。
- TU-net(T2+T1CWI联合输入)性能最佳,阅片时可优先关注联合序列建模。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
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.
引用
- Xu, Xu, Shi, et al. A multitask classification framework based on vision transformer for predicting molecular expressions of glioma. Eur J Radiol. 2022. PMID: 36327857. doi:10.1016/j.ejrad.2022.110560
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。