Department of Radiation Oncology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province, China.
Curr Med Imaging. 2021;17(3):404-409. doi: 10.2174/1573405616666200910141323.
CDATA[Purpose: The aim of this study is to evaluate the accuracy and dosimetric effects for auto- segmentation of the CTV for GO in CT images based on FCN.
An FCN-8s network architecture for auto-segmentation was built based on Caffe. CT images of 121 patients with GO who have received radiotherapy at the West China Hospital of Sichuan University were randomly selected for training and testing. Two methods were used to segment the CTV of GO: treating the two-part CTV as a whole anatomical region or considering the two parts of CTV as two independent regions. Dice Similarity Coefficient (DSC) and Hausdorff Distance (HD) were used as evaluation criteria. The auto-segmented contours were imported into the original treatment plan to analyse the dosimetric characteristics.
The similarity comparison between manual contours and auto-segmental contours showed an average DSC value of up to 0.83. The max HD values for segmenting two parts of CTV separately was a little bit smaller than treating CTV with one label (8.23±2.80 vs. 9.03±2.78). The dosimetric comparison between manual contours and auto-segmental contours showed there was a significant difference (p<0.05) with the lack of dose for auto-segmental CTV.
Based on deep learning architecture, the automatic segmentation model for small target areas can carry out auto contouring tasks well. Treating separate parts of one target as different anatomic regions can help to improve the auto-contouring quality. The dosimetric evaluation can provide us with different perspectives for further exploration of automatic sketching tools.
本研究旨在评估基于 FCN 的自动分割 CT 图像中 GO 的 CTV 的准确性和剂量学影响。
基于 Caffe 构建了用于自动分割的 FCN-8s 网络架构。随机选择了 121 例在四川大学华西医院接受放射治疗的 GO 患者的 CT 图像进行训练和测试。采用两种方法对 GO 的 CTV 进行分割：将两部分 CTV 视为一个整体解剖区域或考虑 CTV 的两部分为两个独立区域。采用 Dice 相似系数（DSC）和 Hausdorff 距离（HD）作为评价标准。将自动分割的轮廓导入原始治疗计划，分析剂量学特征。
手动轮廓与自动分割轮廓之间的相似性比较显示，平均 DSC 值高达 0.83。分别分割两部分 CTV 的最大 HD 值略小于用一个标签处理 CTV（8.23±2.80 比 9.03±2.78）。手动轮廓与自动分割轮廓之间的剂量学比较显示，自动分割 CTV 存在显著差异（p<0.05），剂量不足。
基于深度学习架构，小目标区域的自动分割模型可以很好地完成自动轮廓任务。将一个目标的单独部分视为不同的解剖区域有助于提高自动轮廓质量。剂量学评估可以为我们提供不同的视角，以进一步探索自动勾画工具。
