Department of Radiology, Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Peking University Cancer Hospital & Institute, Beijing, China.
J Appl Clin Med Phys. 2021 Sep;22(9):324-331. doi: 10.1002/acm2.13381. Epub 2021 Aug 3.
Manual delineation of a rectal tumor on a volumetric image is time-consuming and subjective. Deep learning has been used to segment rectal tumors automatically on T2-weighted images, but automatic segmentation on diffusion-weighted imaging is challenged by noise, artifact, and low resolution. In this study, a volumetric U-shaped neural network (U-Net) is proposed to automatically segment rectal tumors on diffusion-weighted images.
Three hundred patients of locally advanced rectal cancer were enrolled in this study and divided into a training group, a validation group, and a test group. The region of rectal tumor was delineated on the diffusion-weighted images by experienced radiologists as the ground truth. A U-Net was designed with a volumetric input of the diffusion-weighted images and an output of segmentation with the same size. A semi-automatic segmentation method was used for comparison by manually choosing a threshold of gray level and automatically selecting the largest connected region. Dice similarity coefficient (DSC) was calculated to evaluate the methods.
On the test group, deep learning method (DSC = 0.675 ± 0.144, median DSC is 0.702, maximum DSC is 0.893, and minimum DSC is 0.297) showed higher segmentation accuracy than the semi-automatic method (DSC = 0.614 ± 0.225, median DSC is 0.685, maximum DSC is 0.869, and minimum DSC is 0.047). Paired t-test shows significant difference (T = 2.160, p = 0.035) in DSC between the deep learning method and the semi-automatic method in the test group.
Volumetric U-Net can automatically segment rectal tumor region on DWI images of locally advanced rectal cancer.
在容积图像上手动勾画直肠肿瘤费时且主观。深度学习已被用于在 T2 加权图像上自动分割直肠肿瘤，但在扩散加权成像上的自动分割受到噪声、伪影和低分辨率的挑战。在本研究中，提出了一种用于在扩散加权图像上自动分割直肠肿瘤的容积 U 形神经网络（U-Net）。
本研究纳入了 300 例局部进展期直肠癌患者，将其分为训练组、验证组和测试组。由经验丰富的放射科医生在扩散加权图像上勾画直肠肿瘤区域作为金标准。设计了一个具有扩散加权图像容积输入和相同大小分割输出的 U-Net。通过手动选择灰度级阈值和自动选择最大连通区域，比较了一种半自动分割方法。计算了 Dice 相似系数（DSC）以评估方法。
在测试组中，深度学习方法（DSC=0.675±0.144，中位数 DSC 为 0.702，最大 DSC 为 0.893，最小 DSC 为 0.297）的分割准确性高于半自动方法（DSC=0.614±0.225，中位数 DSC 为 0.685，最大 DSC 为 0.869，最小 DSC 为 0.047）。配对 t 检验显示，在测试组中，深度学习方法与半自动方法在 DSC 上的差异具有统计学意义（T=2.160，p=0.035）。
容积 U-Net 可自动分割局部进展期直肠癌 DWI 图像上的直肠肿瘤区域。
