[论文解读] Comprehensive Validation of Automated Whole Body Skeletal Muscle, Adipose Tissue, and Bone Segmentation from 3D CT images for Body Composition Analysis: Towards Extended Body Composition

Da Ma, Vincent Chow+2 more|arXiv (Cornell University)|Jun 1, 2021

Nutrition and Health in Aging参考文献 10被引用 14

一句话总结

本文提出了一种基于深度学习的3D全器官分割框架，可对常规3D CT扫描中的骨骼肌、脂肪组织（皮下脂肪和内脏脂肪）及骨骼实现自动化、高精度的定量分析。在50例患者扫描中验证，Dice分数高达0.980–0.986，实现了精确的体积与纹理测量，为精准医学中的扩展体成分分析提供了关键支持。

ABSTRACT

The latest advances in computer-assisted precision medicine are making it feasible to move from population-wide models that are useful to discover aggregate patterns that hold for group-based analysis to patient-specific models that can drive patient-specific decisions with regard to treatment choices, and predictions of outcomes of treatment. Body Composition is recognized as an important driver and risk factor for a wide variety of diseases, as well as a predictor of individual patient-specific clinical outcomes to treatment choices or surgical interventions. 3D CT images are routinely acquired in the oncological worklows and deliver accurate rendering of internal anatomy and therefore can be used opportunistically to assess the amount of skeletal muscle and adipose tissue compartments. Powerful tools of artificial intelligence such as deep learning are making it feasible now to segment the entire 3D image and generate accurate measurements of all internal anatomy. These will enable the overcoming of the severe bottleneck that existed previously, namely, the need for manual segmentation, which was prohibitive to scale to the hundreds of 2D axial slices that made up a 3D volumetric image. Automated tools such as presented here will now enable harvesting whole-body measurements from 3D CT or MRI images, leading to a new era of discovery of the drivers of various diseases based on individual tissue, organ volume, shape, and functional status. These measurements were hitherto unavailable thereby limiting the field to a very small and limited subset. These discoveries and the potential to perform individual image segmentation with high speed and accuracy are likely to lead to the incorporation of these 3D measures into individual specific treatment planning models related to nutrition, aging, chemotoxicity, surgery and survival after the onset of a major disease such as cancer.

研究动机与目标

• 通过实现完整的3D体积分析，克服2D单层面CT测量在预测个体患者预后方面的局限性。
• 开发一种自动化、基于深度学习的分割方法，能够以高精度和高速度处理全器官3D CT扫描。
• 实现对多种组织和器官的全面体成分指标提取——包括体积、形态与纹理特征，服务于精准医学。
• 通过以直接的3D测量替代代理估计，为个体化临床预测模型的建立奠定基础。
• 支持将先进的影像生物标志物整合到癌症、衰老及代谢性疾病患者个体化治疗规划中。

提出的方法

• 采用基于深度学习的3D U-Net架构，对3D CT容积中所有轴向切片实现全器官分割。
• 在包含50例3D CT扫描的多样化数据集上进行端到端训练，标签由专家标注，涵盖骨骼、骨骼肌、皮下脂肪与内脏脂肪。
• 应用数据增强与归一化技术，提升在不同CT扫描协议（对比增强、非对比、低剂量）下的鲁棒性。
• 采用多结构损失函数，结合Dice系数与交叉熵，以优化所有组织类型的分割精度。
• 通过在保留测试集（50例扫描）上使用Dice系数（分割重叠的标准度量）验证性能。
• 为实现临床可扩展性，该方法已集成至商业软件平台（DAFS 3.0），支持真实世界部署。

实验结果

研究问题

• RQ1 基于全器官CT扫描的自动化3D分割能否在骨骼肌、脂肪组织与骨骼的分割中达到足够临床应用的高精度？
• RQ2 3D体积测量与2D单层面（如L3）测量相比，在预测真实全器官组织体积方面表现如何？
• RQ3 基于深度学习的分割在个体化预后预测中，能在多大程度上克服基于回归的代理估计的局限性？
• RQ4 该方法能否在包括非对比、对比增强及低剂量扫描在内的多种CT采集协议中实现良好泛化？
• RQ5 3D体成分指标在超越传统BMI或体表面积（BSA）模型的基础上，对改善个性化治疗规划具有多大潜力？

主要发现

• 在50例CT扫描的测试集中，所提出的3D分割模型对骨骼的平均Dice系数为0.980，对骨骼肌为0.974，对皮下脂肪为0.986，对内脏脂肪组织为0.960。
• 较高的Dice分数表明分割精度接近专家水平，证明了自动化全器官组织定量分析的可行性。
• 该方法在估计真实3D组织体积方面显著优于2D单层面方法，将误差范围从10–30%降低至接近理想精度。
• 在未见的多样化CT扫描（包括对比增强与非对比扫描）中进行的实证测试，证实了其在训练数据之外的强大泛化性能。
• 该框架可从多个器官（肝脏、脾脏、肾脏、胰腺）中提取3D体积、形态与放射组学纹理特征，支持扩展的体成分模型构建。
• 该方法已通过Data Analysis Facilitation Suite 3.0商业化提供，支持临床部署与纵向患者监测。