DataSheet_1_Deep learning auto-segmentation of cervical skeletal muscle for sarcopenia analysis in patients with head and neck cancer.docx

Background/PurposeSarcopenia is a prognostic factor in patients with head and neck cancer (HNC). Sarcopenia can be determined using the skeletal muscle index (SMI) calculated from cervical neck skeletal muscle (SM) segmentations. However, SM segmentation requires manual input, which is time-consuming and variable. Therefore, we developed a fully-automated approach to segment cervical vertebra SM.
Materials/Methods390 HNC patients with contrast-enhanced CT scans were utilized (300-training, 90-testing). Ground-truth single-slice SM segmentations at the C3 vertebra were manually generated. A multi-stage deep learning pipeline was developed, where a 3D ResUNet auto-segmented the C3 section (33 mm window), the middle slice of the section was auto-selected, and a 2D ResUNet auto-segmented the auto-selected slice. Both the 3D and 2D approaches trained five sub-models (5-fold cross-validation) and combined sub-model predictions on the test set using majority vote ensembling. Model performance was primarily determined using the Dice similarity coefficient (DSC). Predicted SMI was calculated using the auto-segmented SM cross-sectional area. Finally, using established SMI cutoffs, we performed a Kaplan-Meier analysis to determine associations with overall survival.
ResultsMean test set DSC of the 3D and 2D models were 0.96 and 0.95, respectively. Predicted SMI had high correlation to the ground-truth SMI in males and females (r>0.96). Predicted SMI stratified patients for overall survival in males (log-rank p = 0.01) but not females (log-rank p = 0.07), consistent with ground-truth SMI.
ConclusionWe developed a high-performance, multi-stage, fully-automated approach to segment cervical vertebra SM. Our study is an essential step towards fully-automated sarcopenia-related decision-making in patients with HNC.

背景与目的：肌肉减少症（Sarcopenia）是头颈癌（head and neck cancer, HNC）患者的预后因素。肌肉减少症可通过基于颈部骨骼肌（skeletal muscle, SM）分割计算得到的骨骼肌指数（skeletal muscle index, SMI）进行判定。然而，骨骼肌分割需依赖人工操作，不仅耗时较长，且结果存在一定变异性。为此，本研究开发了一种全自动的颈椎骨骼肌分割方法。
材料与方法：本研究共纳入390名头颈癌患者的增强CT扫描数据，其中300例用于训练集，90例用于测试集。由人工标注生成第3颈椎（C3）层面的单一层面骨骼肌分割真值（ground-truth）。本研究构建了一套多阶段深度学习流程：首先通过3D ResUNet自动分割C3区域（窗宽设置为33mm），自动选取该区域的中间层面，再通过2D ResUNet对自动选取的层面进行分割。3D与2D两种模型均通过5折交叉验证训练5个子模型，并在测试集上采用多数投票集成（majority vote ensembling）融合各子模型的预测结果。模型性能主要通过戴斯相似性系数（Dice similarity coefficient, DSC）进行评估。利用自动分割得到的骨骼肌横截面积计算预测的骨骼肌指数。最后，基于已确立的骨骼肌指数截断值，开展Kaplan-Meier分析以探究其与患者总生存期的关联。
结果：3D与2D模型在测试集上的平均戴斯相似性系数分别为0.96与0.95。预测得到的骨骼肌指数在男性与女性群体中均与真值骨骼肌指数呈现高度相关性（r>0.96）。预测的骨骼肌指数可对男性患者的总生存期进行分层（log-rank p=0.01），但在女性群体中未观察到显著关联（log-rank p=0.07），该结果与真值骨骼肌指数的分析结果一致。
结论：本研究开发了一种高性能、多阶段的全自动颈椎骨骼肌分割方法。本研究为实现头颈癌患者肌肉减少症相关临床决策的全自动化迈出了关键一步。