Objective: To develop and evaluate a deep radiomics model for clinically significant prostate cancer (csPCa, grade group >= 2) detection and compare its performance to Prostate Imaging Reporting and Data System (PI-RADS) assessment in a multicenter cohort. Materials and Methods: This retrospective study analyzed biparametric (T2W and DW) prostate MRI sequences of 615 patients (mean age, 63.1 +/- 7 years) from four datasets acquired between 2010 and 2020: PROSTATEx challenge, Prostate158 challenge, PCaMAP trial, and an in-house (NTNU/St. Olavs Hospital) dataset. With expert annotations as ground truth, a deep radiomics model was trained, including nnU-Net segmentation of the prostate gland, voxel-wise radiomic feature extraction, extreme gradient boost classification, and post-processing of tumor probability maps into csPCa detection maps. Training involved 5-fold cross-validation using the PROSTATEx (n=199), Prostate158 (n=138), and PCaMAP (n=78) datasets, and testing on the in-house (n=200) dataset. Patient- and lesion-level performance were compared to PI-RADS using area under ROC curve (AUROC [95% CI]), sensitivity, and specificity analysis. Results: On the test data, the radiologist achieved a patient-level AUROC of 0.94 [0.91-0.98] with 94% (75/80) sensitivity and 77% (92/120) specificity at PI-RADS >= 3. The deep radiomics model at a tumor probability cut-off >= 0.76 achieved 0.91 [0.86-0.95] AUROC with 90% (72/80) sensitivity and 73% (87/120) specificity, not significantly different (p = 0.068) from PI-RADS. On the lesion level, PI-RADS cut-off >= 3 had 84% (91/108) sensitivity at 0.2 (40/200) false positives per patient, while deep radiomics attained 68% (73/108) sensitivity at the same false positive rate. Conclusion: Deep radiomics machine learning model achieved comparable performance to PI-RADS assessment in csPCa detection at the patient-level but not at the lesion-level.

翻译：目的：开发并评估一种用于检测临床显著性前列腺癌（csPCa，分级组≥2）的深度影像组学模型，并在多中心队列中将其性能与前列腺影像报告和数据系统（PI-RADS）评估进行比较。材料与方法：这项回顾性研究分析了来自四个数据集（采集于2010年至2020年间）的615名患者（平均年龄63.1±7岁）的双参数（T2W和DW）前列腺MRI序列：PROSTATEx挑战赛、Prostate158挑战赛、PCaMAP试验以及一个内部（挪威科技大学/圣奥拉夫医院）数据集。以专家标注为金标准，训练了一个深度影像组学模型，包括使用nnU-Net分割前列腺、逐体素影像组学特征提取、极限梯度提升分类，以及将肿瘤概率图后处理为csPCa检测图。训练采用5折交叉验证，使用PROSTATEx（n=199）、Prostate158（n=138）和PCaMAP（n=78）数据集，并在内部（n=200）数据集上进行测试。使用受试者工作特征曲线下面积（AUROC [95% CI]）、敏感性和特异性分析，在患者层面和病灶层面将模型性能与PI-RADS进行比较。结果：在测试数据上，放射科医生在患者层面（PI-RADS ≥ 3）的AUROC为0.94 [0.91-0.98]，敏感性为94%（75/80），特异性为77%（92/120）。深度影像组学模型在肿瘤概率截断值≥0.76时，AUROC为0.91 [0.86-0.95]，敏感性为90%（72/80），特异性为73%（87/120），与PI-RADS无显著差异（p = 0.068）。在病灶层面，PI-RADS截断值≥3的敏感性为84%（91/108），每例患者假阳性数为0.2（40/200）；而深度影像组学在相同假阳性率下的敏感性为68%（73/108）。结论：深度影像组学机器学习模型在患者层面的csPCa检测中取得了与PI-RADS评估相当的性能，但在病灶层面则不然。