de Almeida José Guilherme, Rodrigues Nuno M, Castro Verde Ana Sofia, Mascarenhas Gaivão Ana, Bilreiro Carlos, Santiago Inês, Ip Joana, Belião Sara, Matos Celso, Silva Sara, Tsiknakis Manolis, Marias Kostantinos, Regge Daniele, Papanikolaou Nikolaos
Champalimaud Research, Champalimaud Foundation, Avenida Brasilia, Lisboa, Lisboa 1400-038 Portugal.
LASIGE, Faculdade de Ciências, Universidade de Lisboa, Portugal.
Radiol Artif Intell. 2025 May;7(3):e230555. doi: 10.1148/ryai.230555.
Purpose To assess the effect of scanner manufacturer and scanning protocol on the performance of deep learning models to classify aggressiveness of prostate cancer (PCa) at biparametric MRI (bpMRI). Materials and Methods In this retrospective study, 5478 cases from ProstateNet, a PCa bpMRI dataset with examinations from 13 centers, were used to develop five deep learning (DL) models to predict PCa aggressiveness with minimal lesion information and test how using data from different subgroups-scanner manufacturers and endorectal coil (ERC) use (Siemens, Philips, GE with and without ERC, and the full dataset)-affects model performance. Performance was assessed using the area under the receiver operating characteristic curve (AUC). The effect of clinical features (age, prostate-specific antigen level, Prostate Imaging Reporting and Data System score) on model performance was also evaluated. Results DL models were trained on 4328 bpMRI cases, and the best model achieved an AUC of 0.73 when trained and tested using data from all manufacturers. Held-out test set performance was higher when models trained with data from a manufacturer were tested on the same manufacturer (within- and between-manufacturer AUC differences of 0.05 on average, < .001). The addition of clinical features did not improve performance ( = .24). Learning curve analyses showed that performance remained stable as training data increased. Analysis of DL features showed that scanner manufacturer and scanning protocol heavily influenced feature distributions. Conclusion In automated classification of PCa aggressiveness using bpMRI data, scanner manufacturer and ERC use had a major effect on DL model performance and features. Convolutional Neural Network (CNN), Computer-aided Diagnosis (CAD), Computer Applications-General (Informatics), Oncology Published under a CC BY 4.0 license. See also commentary by Suri and Hsu in this issue.
目的 评估扫描仪制造商和扫描协议对深度学习模型在双参数磁共振成像（bpMRI）中对前列腺癌（PCa）侵袭性进行分类的性能的影响。材料与方法 在这项回顾性研究中，使用了来自ProstateNet的5478例病例，这是一个来自13个中心检查的PCa的bpMRI数据集，用于开发五个深度学习（DL）模型，以利用最少的病变信息预测PCa侵袭性，并测试使用来自不同子组的数据——扫描仪制造商和直肠内线圈（ERC）的使用情况（西门子、飞利浦、通用电气，使用和不使用ERC，以及完整数据集）——如何影响模型性能。使用受试者操作特征曲线下面积（AUC）评估性能。还评估了临床特征（年龄、前列腺特异性抗原水平、前列腺影像报告和数据系统评分）对模型性能的影响。结果 DL模型在4328例bpMRI病例上进行训练，当使用来自所有制造商的数据进行训练和测试时，最佳模型的AUC达到0.73。当用来自某一制造商的数据训练的模型在同一制造商的数据上进行测试时，留出测试集的性能更高（制造商内部和制造商之间的AUC差异平均为0.05，P <.001）。添加临床特征并未提高性能（P =.24）。学习曲线分析表明，随着训练数据的增加，性能保持稳定。对DL特征的分析表明，扫描仪制造商和扫描协议对特征分布有重大影响。结论 在使用bpMRI数据自动分类PCa侵袭性时，扫描仪制造商和ERC的使用对DL模型性能和特征有重大影响。卷积神经网络（CNN）、计算机辅助诊断（CAD）、计算机应用——一般（信息学）、肿瘤学 本研究根据知识共享署名4.0许可协议发布。另见本期Suri和Hsu的评论。
