Bao Jie, Zhao Litao, Qiao Xiaomeng, Li Zhenkai, Ji Yanting, Su Yueting, Ji Libiao, Shen Junkang, Liu Jiangang, Tian Jie, Wang Ximing, Shen Hailin, Hu Chunhong
Department of Radiology, The First Affiliated Hospital of Soochow University, Suzhou, China.
School of Engineering Medicine, Beihang University, Beijing, China.
Insights Imaging. 2025 Jan 29;16(1):25. doi: 10.1186/s13244-024-01896-1.
The presence of clinically significant prostate cancer (csPCa) is equivocal for patients with prostate imaging reporting and data system (PI-RADS) category 3. We aim to develop deep learning models for re-stratify risks in PI-RADS category 3 patients.
This retrospective study included a bi-parametric MRI of 1567 consecutive male patients from six centers (Centers 1-6) between Jan 2015 and Dec 2020. Deep learning models with double channel attention modules based on MRI (AttenNet) for predicting PCa and csPCa were constructed separately. Each model was first pretrained using 1144 PI-RADS 1-2 and 4-5 images and then retrained using 238 PI-RADS 3 images from three training centers (centers 1-3), and tested using 185 PI-RADS 3 images from the other three testing centers (centers 4-6).
Our AttenNet models achieved excellent prediction performances in testing cohort of center 4-6 with the area under the receiver operating characteristic curves (AUC) of 0.795 (95% CI: [0.700, 0.891]), 0.963 (95% CI: [0.915, 1]) and 0.922 (95% CI: [0.810, 1]) in predicting PCa, and the corresponding AUCs were 0.827 (95% CI: [0.703, 0.952]) and 0.926 (95% CI: [0.846, 1]) in predicting csPCa in testing cohort of center 4 and center 5. In particular, 71.1% to 92.2% of non-csPCa patients were identified by our model in three testing cohorts, who can spare from invasive biopsy or RP procedure.
Our model offers a noninvasive screening clinical tool to re-stratify risks in PI-RADS 3 patients, thereby reducing unnecessary invasive biopsies and improving the effectiveness of biopsies.
The deep learning model with MRI can help to screen out csPCa in PI-RADS category 3.
AttenNet models included channel attention and soft attention modules. 71.1-92.2% of non-csPCa patients were identified by the AttenNet model. The AttenNet models can be a screen clinical tool to re-stratify risks in PI-RADS 3 patients.
对于前列腺影像报告和数据系统（PI-RADS）分类为3类的患者，临床显著前列腺癌（csPCa）的存在情况尚不明确。我们旨在开发深度学习模型，对PI-RADS分类为3类的患者重新进行风险分层。
这项回顾性研究纳入了2015年1月至2020年12月期间来自六个中心（中心1 - 6）的1567名男性患者的双参数MRI数据。分别构建基于MRI的带有双通道注意力模块的深度学习模型（AttenNet）来预测前列腺癌（PCa）和csPCa。每个模型首先使用1144张PI-RADS 1 - 2类和4 - 5类图像进行预训练，然后使用来自三个训练中心（中心1 - 3）的238张PI-RADS 3类图像进行再训练，并使用来自其他三个测试中心（中心4 - 6）的185张PI-RADS 3类图像进行测试。
我们的AttenNet模型在中心4 - 6的测试队列中取得了优异的预测性能，在预测PCa时，受试者工作特征曲线下面积（AUC）分别为0.795（95%置信区间：[0.700, 0.891]）、0.963（95%置信区间：[0.915, 1]）和0.922（95%置信区间：[0.810, 1]）；在中心4和中心5的测试队列中预测csPCa时，相应的AUC分别为0.827（95%置信区间：[0.703, 0.952]）和0.926（95%置信区间：[0.846, 1]）。特别是，在三个测试队列中，我们的模型识别出了71.1%至92.2%的非csPCa患者，这些患者可避免进行侵入性活检或根治性前列腺切除术（RP）。
我们的模型提供了一种非侵入性筛查临床工具，用于对PI-RADS 3类患者重新进行风险分层，从而减少不必要的侵入性活检并提高活检的有效性。
带有MRI的深度学习模型有助于筛查出PI-RADS分类为3类的csPCa。
AttenNet模型包括通道注意力和软注意力模块。AttenNet模型识别出了71.1 - 92.2%的非csPCa患者。AttenNet模型可以作为一种筛查临床工具，对PI-RADS 3类患者重新进行风险分层。
