深度学习基础的临床靶区三通道自动分割算法在宫颈癌自适应放疗中的临床评估。
Clinical evaluation of deep learning-based clinical target volume three-channel auto-segmentation algorithm for adaptive radiotherapy in cervical cancer.
机构信息
Department of Radiation Oncology, 1st Affiliated Hospital of Soochow University, No. 188 Shizi Street, Suzhou, 215123, China.
Shanghai United Imaging Healthcare, Co. Ltd., Jiading, 201807, China.
出版信息
BMC Med Imaging. 2022 Jul 9;22(1):123. doi: 10.1186/s12880-022-00851-0.
OBJECTIVES
Accurate contouring of the clinical target volume (CTV) is a key element of radiotherapy in cervical cancer. We validated a novel deep learning (DL)-based auto-segmentation algorithm for CTVs in cervical cancer called the three-channel adaptive auto-segmentation network (TCAS).
METHODS
A total of 107 cases were collected and contoured by senior radiation oncologists (ROs). Each case consisted of the following: (1) contrast-enhanced CT scan for positioning, (2) the related CTV, (3) multiple plain CT scans during treatment and (4) the related CTV. After registration between (1) and (3) for the same patient, the aligned image and CTV were generated. Method 1 is rigid registration, method 2 is deformable registration, and the aligned CTV is seen as the result. Method 3 is rigid registration and TCAS, method 4 is deformable registration and TCAS, and the result is generated by a DL-based method.
RESULTS
From the 107 cases, 15 pairs were selected as the test set. The dice similarity coefficient (DSC) of method 1 was 0.8155 ± 0.0368; the DSC of method 2 was 0.8277 ± 0.0315; the DSCs of method 3 and 4 were 0.8914 ± 0.0294 and 0.8921 ± 0.0231, respectively. The mean surface distance and Hausdorff distance of methods 3 and 4 were markedly better than those of method 1 and 2.
CONCLUSIONS
The TCAS achieved comparable accuracy to the manual delineation performed by senior ROs and was significantly better than direct registration.
目的
精确勾画临床靶区（CTV）是宫颈癌放射治疗的关键要素。我们验证了一种名为三通道自适应自动分割网络（TCAS）的新型基于深度学习（DL）的宫颈癌CTV 自动分割算法。
方法
共收集了 107 例患者的资料，由资深放射肿瘤学家（RO）进行勾画。每个病例包括以下内容：（1）用于定位的增强 CT 扫描，（2）相关 CTV，（3）治疗期间的多个平扫 CT 扫描，（4）相关 CTV。在同一患者的（1）和（3）之间进行注册后，生成配准图像和 CTV。方法 1 是刚性注册，方法 2 是变形注册，配准 CTV 为结果。方法 3 是刚性注册和 TCAS，方法 4 是变形注册和 TCAS，DL 方法生成结果。
结果
从 107 例中选择了 15 对作为测试集。方法 1 的骰子相似系数（DSC）为 0.8155±0.0368；方法 2 的 DSC 为 0.8277±0.0315；方法 3 和 4 的 DSCs 分别为 0.8914±0.0294 和 0.8921±0.0231。方法 3 和 4 的平均表面距离和 Hausdorff 距离明显优于方法 1 和 2。
结论
TCAS 达到了与资深 RO 手动勾画相当的准确性，并且明显优于直接注册。
Zotero 插件