Shanghai Key Laboratory of Magnetic Resonance, East China Normal University, Shanghai, China.
Department of Radiology, First Affiliated Hospital with Nanjing Medical University, Nanjing, China.
J Magn Reson Imaging. 2018 Dec;48(6):1570-1577. doi: 10.1002/jmri.26047. Epub 2018 Apr 16.
BACKGROUND: Deep learning is the most promising methodology for automatic computer-aided diagnosis of prostate cancer (PCa) with multiparametric MRI (mp-MRI). PURPOSE: To develop an automatic approach based on deep convolutional neural network (DCNN) to classify PCa and noncancerous tissues (NC) with mp-MRI. STUDY TYPE: Retrospective. SUBJECTS: In all, 195 patients with localized PCa were collected from a PROSTATEx database. In total, 159/17/19 patients with 444/48/55 observations (215/23/23 PCas and 229/25/32 NCs) were randomly selected for training/validation/testing, respectively. SEQUENCE: T -weighted, diffusion-weighted, and apparent diffusion coefficient images. ASSESSMENT: A radiologist manually labeled the regions of interest of PCas and NCs and estimated the Prostate Imaging Reporting and Data System (PI-RADS) scores for each region. Inspired by VGG-Net, we designed a patch-based DCNN model to distinguish between PCa and NCs based on a combination of mp-MRI data. Additionally, an enhanced prediction method was used to improve the prediction accuracy. The performance of DCNN prediction was tested using a receiver operating characteristic (ROC) curve, and the area under the ROC curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. Moreover, the predicted result was compared with the PI-RADS score to evaluate its clinical value using decision curve analysis. STATISTICAL TEST: Two-sided Wilcoxon signed-rank test with statistical significance set at 0.05. RESULTS: The DCNN produced excellent diagnostic performance in distinguishing between PCa and NC for testing datasets with an AUC of 0.944 (95% confidence interval: 0.876-0.994), sensitivity of 87.0%, specificity of 90.6%, PPV of 87.0%, and NPV of 90.6%. The decision curve analysis revealed that the joint model of PI-RADS and DCNN provided additional net benefits compared with the DCNN model and the PI-RADS scheme. DATA CONCLUSION: The proposed DCNN-based model with enhanced prediction yielded high performance in statistical analysis, suggesting that DCNN could be used in computer-aided diagnosis (CAD) for PCa classification. LEVEL OF EVIDENCE: 3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018;48:1570-1577.
背景：深度学习是一种最有前途的方法，可用于对前列腺癌（PCa）的多参数 MRI（mp-MRI）进行自动计算机辅助诊断。
目的：开发一种基于深度卷积神经网络（DCNN）的自动方法，用于对 mp-MRI 中的 PCa 和非癌组织（NC）进行分类。
研究类型：回顾性。
受试者：共从 PROSTATEx 数据库中收集了 195 例局限性 PCa 患者。总共随机选择了 159/17/19 例患者的 444/48/55 个观察结果（215/23/23 例 PCa 和 229/25/32 例 NC）进行训练/验证/测试。
序列：T 加权、弥散加权和表观弥散系数图像。
评估：一位放射科医生手动标记了 PCa 和 NC 的感兴趣区域，并为每个区域估计了前列腺成像报告和数据系统（PI-RADS）评分。受 VGG-Net 的启发，我们设计了一种基于补丁的 DCNN 模型，该模型基于 mp-MRI 数据组合来区分 PCa 和 NC。此外，还使用增强预测方法来提高预测精度。使用受试者工作特征（ROC）曲线测试 DCNN 预测的性能，并计算 ROC 曲线下的面积（AUC）、敏感性、特异性、阳性预测值（PPV）和阴性预测值（NPV）。此外，还通过决策曲线分析将预测结果与 PI-RADS 评分进行比较，以评估其临床价值。
统计检验：采用双侧 Wilcoxon 符号秩检验，显著性水平为 0.05。
结果：DCNN 在测试数据集上区分 PCa 和 NC 的诊断性能非常出色，AUC 为 0.944（95%置信区间：0.876-0.994），敏感性为 87.0%，特异性为 90.6%，PPV 为 87.0%，NPV 为 90.6%。决策曲线分析表明，PI-RADS 和 DCNN 的联合模型与 DCNN 模型和 PI-RADS 方案相比提供了额外的净收益。
数据结论：提出的基于 DCNN 的增强预测模型在统计学分析中表现出了较高的性能，表明 DCNN 可用于 PCa 分类的计算机辅助诊断（CAD）。
证据水平：3 级 磁共振成像杂志 2018 年；48 期：1570-1577 页
J Magn Reson Imaging. 2018-4-16
J Magn Reson Imaging. 2018-9-19
J Magn Reson Imaging. 2018-10-22
J Magn Reson Imaging. 2018-2-13
J Magn Reson Imaging. 2019-2-27
Bioengineering (Basel). 2023-8-28
Zotero 插件