Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, United States.
Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, United States.
Radiother Oncol. 2021 Jun;159:1-7. doi: 10.1016/j.radonc.2021.02.040. Epub 2021 Mar 3.
Artificial intelligence advances have stimulated a new generation of autosegmentation, however clinical evaluations of these algorithms are lacking. This study assesses the clinical utility of deep learning-based autosegmentation for MR-based prostate radiotherapy planning.
Data was collected prospectively for patients undergoing prostate-only radiation at our institution from June to December 2019. Geometric indices (volumetric Dice-Sørensen Coefficient, VDSC; surface Dice-Sørensen Coefficient, SDSC; added path length, APL) compared automated to final contours. Physicians reported contouring time and rated autocontours on 3-point protocol deviation scales. Descriptive statistics and univariable analyses evaluated relationships between the aforementioned metrics.
Among 173 patients, 85% received SBRT. The CTV was available for 167 (97%) with median VDSC, SDSC, and APL for CTV (prostate and SV) 0.89 (IQR 0.83-0.95), 0.91 (IQR 0.75-0.96), and 1801 mm (IQR 1140-2703), respectively. Physicians completed surveys for 43/55 patients (RR 78%). 33% of autocontours (14/43) required major "clinically significant" edits. Physicians spent a median of 28 min contouring (IQR 20-30), representing a 12-minute (30%) time savings compared to historic controls (median 40, IQR 25-68, n = 21, p < 0.01). Geometric indices correlated weakly with contouring time, and had no relationship with quality scores.
Deep learning-based autosegmentation was implemented successfully and improved efficiency. Major "clinically significant" edits are uncommon and do not correlate with geometric indices. APL was supported as a clinically meaningful quantitative metric. Efforts are needed to educate and generate consensus among physicians, and develop mechanisms to flag cases for quality assurance.
人工智能的进步激发了新一代的自动分割，但这些算法的临床评估却很缺乏。本研究评估了基于深度学习的自动分割在基于磁共振的前列腺放射治疗计划中的临床应用价值。
我们从 2019 年 6 月至 12 月前瞻性地收集了在我们机构接受前列腺放射治疗的患者的数据。几何指标（体积 Dice-Sørensen 系数，VDSC；表面 Dice-Sørensen 系数，SDSC；附加路径长度，APL）比较了自动和最终轮廓。医生报告了勾画时间，并使用 3 分协议偏离量表对自动轮廓进行了评分。描述性统计和单变量分析评估了上述指标之间的关系。
在 173 名患者中，85%接受了 SBRT。167 名患者（97%）的 CTV 可用，CTV（前列腺和 SV）的中位 VDSC、SDSC 和 APL 分别为 0.89（IQR 0.83-0.95）、0.91（IQR 0.75-0.96）和 1801mm（IQR 1140-2703）。医生完成了 55 名患者中 43 名（RR 78%）的调查。33%的自动轮廓（14/43）需要进行主要的“临床显著”编辑。医生的勾画时间中位数为 28 分钟（IQR 20-30），与历史对照（中位数 40 分钟，IQR 25-68 分钟，n=21，p<0.01）相比，时间节省了 12 分钟（30%）。几何指标与勾画时间呈弱相关，与质量评分无关。
基于深度学习的自动分割成功实施并提高了效率。主要的“临床显著”编辑并不常见，与几何指标无关。APL 被认为是一种有临床意义的定量指标。需要努力教育和达成医生共识，并开发机制来标记质量保证案例。
