Institute of Radiation Physics, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland.
Radiation Oncology Department, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland.
Phys Med. 2024 Jul;123:103402. doi: 10.1016/j.ejmp.2024.103402. Epub 2024 Jun 12.
One of the advantages of integrating automated processes in treatment planning is the reduction of manual planning variability. This study aims to assess whether a deep-learning-based auto-planning solution can also reduce the contouring variation-related impact on the planned dose for early-breast cancer treatment.
Auto- and manual plans were optimized for 20 patients using both auto- and manual OARs, including both lungs, right breast, heart, and left-anterior-descending (LAD) artery. Differences in terms of recalculated dose (ΔD,ΔD) and reoptimized dose (ΔD,ΔD) for manual (M) and auto (A)-plans, were evaluated on manual structures. The correlation between several geometric similarities and dose differences was also explored (Spearman's test).
Auto-contours were found slightly smaller in size than manual contours for right breast and heart and more than twice larger for LAD. Recalculated dose differences were found negligible for both planning approaches except for heart (ΔD=-0.4 Gy, ΔD=-0.3 Gy) and right breast (ΔD=-1.2 Gy, ΔD=-1.3 Gy) maximum dose. Re-optimized dose differences were considered equivalent to recalculated ones for both lungs and LAD, while they were significantly smaller for heart (ΔD=-0.2 Gy, ΔD=-0.2 Gy) and right breast (ΔD =-0.3 Gy, ΔD=-0.9 Gy) maximum dose. Twenty-one correlations were found for ΔD (M=8,A=13) that reduced to four for ΔD (M=3,A=1).
The sensitivity of auto-planning to contouring variation was found not relevant when compared to manual planning, regardless of the method used to calculate the dose differences. Nonetheless, the method employed to define the dose differences strongly affected the correlation analysis resulting highly reduced when dose was reoptimized, regardless of the planning approach.
在治疗计划中集成自动化流程的优势之一是减少手动计划的可变性。本研究旨在评估基于深度学习的自动规划解决方案是否也可以减少与轮廓变化相关的对早期乳腺癌治疗计划剂量的影响。
使用自动和手动 OAR 为 20 名患者优化自动和手动计划，包括双肺、右乳房、心脏和左前降支（LAD）动脉。评估手动结构上手动（M）和自动（A）计划的重新计算剂量（ΔD，ΔD）和重新优化剂量（ΔD，ΔD）之间的差异。还探索了几个几何相似性和剂量差异之间的相关性（Spearman 检验）。
与手动轮廓相比，自动轮廓的右乳房和心脏略小，而 LAD 则大两倍以上。两种计划方法的重新计算剂量差异可忽略不计，除了心脏（ΔD=-0.4Gy，ΔD=-0.3Gy）和右乳房（ΔD=-1.2Gy，ΔD=-1.3Gy）最大剂量。重新优化的剂量差异与重新计算的剂量差异相当，对于双肺和 LAD，而对于心脏（ΔD=-0.2Gy，ΔD=-0.2Gy）和右乳房（ΔD=-0.3Gy，ΔD=-0.9Gy）最大剂量，它们明显更小。对于ΔD（M=8，A=13）发现了 21 个相关性，对于ΔD（M=3，A=1）减少到 4 个。
与手动计划相比，自动计划对轮廓变化的敏感性被认为不相关，无论使用哪种方法计算剂量差异。尽管如此，用于定义剂量差异的方法强烈影响相关分析，当剂量重新优化时，无论采用何种计划方法，相关性都会大大降低。
