Department of Radiation Oncology, Loyola University Chicago, Stritch School of Medicine, Maywood, Illinois, USA.
Department of Radiation Oncology, Cardinal Bernard Cancer Center, Maywood, Illinois, USA.
J Appl Clin Med Phys. 2024 Oct;25(10):e14461. doi: 10.1002/acm2.14461. Epub 2024 Aug 2.
The accuracy of artificial intelligence (AI) generated contours for intact-breast and post-mastectomy radiotherapy plans was evaluated. Geometric and dosimetric comparisons were performed between auto-contours (ACs) and manual-contours (MCs) produced by physicians for target structures. Breast and regional nodal structures were manually delineated on 66 breast cancer patients. ACs were retrospectively generated. The characteristics of the breast/post-mastectomy chestwall (CW) and regional nodal structures (axillary [AxN], supraclavicular [SC], internal mammary [IM]) were geometrically evaluated by Dice similarity coefficient (DSC), mean surface distance, and Hausdorff Distance. The structures were also evaluated dosimetrically by superimposing the MC clinically delivered plans onto the ACs to assess the impact of utilizing ACs with target dose (Vx%) evaluation. Positive geometric correlations between volume and DSC for intact-breast, AxN, and CW were observed. Little or anti correlations between volume and DSC for IM and SC were shown. For intact-breast plans, insignificant dosimetric differences between ACs and MCs were observed for AxN (p = 0.17) and SC (p = 0.16), while IMN ACs and MCs were significantly different. The average V95% for intact-breast MCs (98.4%) and ACs (97.1%) were comparable but statistically different (p = 0.02). For post-mastectomy plans, AxN (p = 0.35) and SC (p = 0.08) were consistent between ACs and MCs, while IMN was significantly different. Additionally, 94.1% of AC-breasts met ΔV95% variation <5% when DSC > 0.7. However, only 62.5% AC-CWs achieved the same metrics, despite AC-CW (p = 0.43) being statistically insignificant. The AC intact-breast structure was dosimetrically similar to MCs. The AC AxN and SC may require manual adjustments. Careful review should be performed for AC post-mastectomy CW and IMN before treatment planning. The findings of this study may guide the clinical decision-making process for the utilization of AI-driven ACs for intact-breast and post-mastectomy plans. Before clinical implementation of this auto-segmentation software, an in-depth assessment of agreement with each local facilities MCs is needed.
评估了人工智能（AI）生成的完整乳房和乳房切除术后放疗计划轮廓的准确性。对目标结构的医生生成的自动轮廓（AC）和手动轮廓（MC）进行了几何和剂量学比较。对 66 例乳腺癌患者的乳房和区域淋巴结结构进行了手动勾画。回顾性生成 AC。通过 Dice 相似系数（DSC）、平均表面距离和 Hausdorff 距离对乳房/乳房切除术后胸壁（CW）和区域淋巴结结构（腋窝 [AxN]、锁骨上 [SC]、内乳 [IM]）的特征进行了几何评估。还通过将 MC 临床交付计划叠加到 AC 上来评估结构的剂量学，以评估利用目标剂量（Vx%）评估的 AC 的影响。完整乳房、AxN 和 CW 的体积与 DSC 之间存在正几何相关性。IM 和 SC 的体积与 DSC 之间几乎没有相关性或呈负相关。对于完整乳房计划，AxN（p=0.17）和 SC（p=0.16）的 AC 和 MC 之间观察到剂量学差异无统计学意义，而 IMN 的 AC 和 MC 之间有显著差异。完整乳房 MCs（98.4%）和 ACs（97.1%）的平均 V95%相似，但有统计学差异（p=0.02）。对于乳房切除术后计划，AxN（p=0.35）和 SC（p=0.08）在 AC 和 MC 之间一致，而 IMN 则有显著差异。此外，当 DSC>0.7 时，94.1%的 AC 乳房满足ΔV95%变化<5%。然而，只有 62.5%的 AC-CW 达到了相同的指标，尽管 AC-CW（p=0.43）没有统计学意义。AC 完整乳房结构与 MC 剂量学相似。AC-AxN 和 SC 可能需要手动调整。在进行治疗计划之前，应仔细审查 AC 乳房切除术后 CW 和 IMN。这项研究的结果可能为完整乳房和乳房切除术后计划中使用 AI 驱动的 AC 提供临床决策过程的指导。在临床实施此自动分割软件之前，需要对其与每个当地设施 MC 的一致性进行深入评估。
