DataSheet_1_Automatic renal mass segmentation and classification on CT images based on 3D U-Net and ResNet algorithms.docx

PurposeTo automatically evaluate renal masses in CT images by using a cascade 3D U-Net- and ResNet-based method to accurately segment and classify focal renal lesions.
Material and MethodsWe used an institutional dataset comprising 610 CT image series from 490 patients from August 2009 to August 2021 to train and evaluate the proposed method. We first determined the boundaries of the kidneys on the CT images utilizing a 3D U-Net-based method to be used as a region of interest to search for renal mass. An ensemble learning model based on 3D U-Net was then used to detect and segment the masses, followed by a ResNet algorithm for classification. Our algorithm was evaluated with an external validation dataset and kidney tumor segmentation (KiTS21) challenge dataset.
ResultsThe algorithm achieved a Dice similarity coefficient (DSC) of 0.99 for bilateral kidney boundary segmentation in the test set. The average DSC for renal mass delineation using the 3D U-Net was 0.75 and 0.83. Our method detected renal masses with recalls of 84.54% and 75.90%. The classification accuracy in the test set was 86.05% for masses (<5 mm) and 91.97% for masses (≥5 mm).
ConclusionWe developed a deep learning-based method for fully automated segmentation and classification of renal masses in CT images. Testing of this algorithm showed that it has the capability of accurately localizing and classifying renal masses.

研究目的：采用基于级联3D U-Net与ResNet的方法，实现CT图像中肾脏肿物的自动化评估，以精准分割并分类局灶性肾脏病变。
材料与方法：本研究纳入2009年8月至2021年8月间，来自490名患者的610例CT影像序列作为机构内部数据集，用于训练与评估所提出的方法。首先，我们利用基于3D U-Net的方法确定CT图像中肾脏的边界，将其作为搜索肾脏肿物的感兴趣区域（region of interest, ROI）；随后，采用基于3D U-Net的集成学习模型完成肿物的检测与分割，再通过ResNet算法完成分类。本研究的算法通过外部验证数据集以及肾脏肿瘤分割挑战赛2021（Kidney Tumor Segmentation Challenge 2021, KiTS21）数据集进行评估。
结果：在测试集上，本算法对双侧肾脏边界分割的戴斯相似性系数（Dice Similarity Coefficient, DSC）达到0.99。基于3D U-Net的肾脏肿物勾画的平均戴斯相似性系数分别为0.75与0.83。本算法检测肾脏肿物的召回率分别为84.54%与75.90%。针对测试集内肿物的分类准确率：直径小于5mm的肿物为86.05%，直径大于等于5mm的肿物为91.97%。
结论：本研究开发了一种基于深度学习的方法，可实现CT图像中肾脏肿物的全自动化分割与分类。算法测试结果表明，其能够精准定位并分类肾脏肿物。