Priester Alan, Fan Richard E, Shubert Joshua, Rusu Mirabela, Vesal Sulaiman, Shao Wei, Khandwala Yash Samir, Marks Leonard S, Natarajan Shyam, Sonn Geoffrey A
Department of Urology, David Geffen School of Medicine, Los Angeles, CA, USA.
Avenda Health, Inc., Culver City, CA, USA.
Eur Urol Open Sci. 2023 Jun 13;54:20-27. doi: 10.1016/j.euros.2023.05.018. eCollection 2023 Aug.
Magnetic resonance imaging (MRI) underestimation of prostate cancer extent complicates the definition of focal treatment margins.
To validate focal treatment margins produced by an artificial intelligence (AI) model.
Testing was conducted retrospectively in an independent dataset of 50 consecutive patients who had radical prostatectomy for intermediate-risk cancer. An AI deep learning model incorporated multimodal imaging and biopsy data to produce three-dimensional cancer estimation maps and margins. AI margins were compared with conventional MRI regions of interest (ROIs), 10-mm margins around ROIs, and hemigland margins. The AI model also furnished predictions of negative surgical margin probability, which were assessed for accuracy.
Comparing AI with conventional margins, sensitivity was evaluated using Wilcoxon signed-rank tests and negative margin rates using chi-square tests. Predicted versus observed negative margin probability was assessed using linear regression. Clinically significant prostate cancer (International Society of Urological Pathology grade ≥2) delineated on whole-mount histopathology served as ground truth.
The mean sensitivity for cancer-bearing voxels was higher for AI margins (97%) than for conventional ROIs (37%, < 0.001), 10-mm ROI margins (93%, = 0.24), and hemigland margins (94%, < 0.001). For index lesions, AI margins were more often negative (90%) than conventional ROIs (0%, < 0.001), 10-mm ROI margins (82%, = 0.24), and hemigland margins (66%, = 0.004). Predicted and observed negative margin probabilities were strongly correlated (R = 0.98, median error = 4%). Limitations include a validation dataset derived from a single institution's prostatectomy population.
The AI model was accurate and effective in an independent test set. This approach could improve and standardize treatment margin definition, potentially reducing cancer recurrence rates. Furthermore, an accurate assessment of negative margin probability could facilitate informed decision-making for patients and physicians.
Artificial intelligence was used to predict the extent of tumors in surgically removed prostate specimens. It predicted tumor margins more accurately than conventional methods.
磁共振成像（MRI）对前列腺癌范围的低估使局灶性治疗切缘的定义变得复杂。
验证人工智能（AI）模型生成的局灶性治疗切缘。
设计、设置和参与者：在一个独立的数据集中对50例因中度风险癌症接受根治性前列腺切除术的连续患者进行回顾性测试。一个AI深度学习模型整合了多模态成像和活检数据，以生成三维癌症估计图和切缘。将AI切缘与传统MRI感兴趣区域（ROI）、ROI周围10毫米切缘以及半腺体切缘进行比较。AI模型还提供了阴性手术切缘概率的预测，并对其准确性进行了评估。
将AI与传统切缘进行比较，使用Wilcoxon符号秩检验评估敏感性，使用卡方检验评估阴性切缘率。使用线性回归评估预测的与观察到的阴性切缘概率。在全切片组织病理学上确定的具有临床意义的前列腺癌（国际泌尿病理学会分级≥2级）作为金标准。
对于含有癌症的体素，AI切缘的平均敏感性（97%）高于传统ROI（37%，<0.001）、10毫米ROI切缘（93%，=0.24）和半腺体切缘（94%，<0.001）。对于索引病变，AI切缘更常为阴性（90%），高于传统ROI（0%，<0.001）、10毫米ROI切缘（82%，=0.24）和半腺体切缘（66%，=0.004）。预测的和观察到的阴性切缘概率高度相关（R=0.98，中位误差=4%）。局限性包括验证数据集来自单一机构的前列腺切除术人群。
AI模型在独立测试集中准确有效。这种方法可以改进和标准化治疗切缘的定义，有可能降低癌症复发率。此外，对阴性切缘概率的准确评估可以促进患者和医生做出明智的决策。
使用人工智能预测手术切除的前列腺标本中的肿瘤范围。它比传统方法更准确地预测了肿瘤切缘。
