[论文解读] A Deep Learning-Based Method for Automatic Segmentation of Proximal Femur from Quantitative Computed Tomography Images

Zhao Chen, Joyce H. Keyak+11 more|arXiv (Cornell University)|Jun 9, 2020

Bone health and osteoporosis research参考文献 14被引用 7

一句话总结

本研究提出了一种基于3D V-Net的深度学习方法，用于在定量计算机断层扫描（QCT）图像中自动分割近端股骨，采用结合Dice损失和L2正则化的复合损失函数。该方法在Dice分数上达到0.9815，决定系数R²为0.9956，显示出高精度和在骨质疏松症及骨折风险评估方面的强大临床潜力。

ABSTRACT

Purpose: Proximal femur image analyses based on quantitative computed tomography (QCT) provide a method to quantify the bone density and evaluate osteoporosis and risk of fracture. We aim to develop a deep-learning-based method for automatic proximal femur segmentation. Methods and Materials: We developed a 3D image segmentation method based on V-Net, an end-to-end fully convolutional neural network (CNN), to extract the proximal femur QCT images automatically. The proposed V-net methodology adopts a compound loss function, which includes a Dice loss and a L2 regularizer. We performed experiments to evaluate the effectiveness of the proposed segmentation method. In the experiments, a QCT dataset which included 397 QCT subjects was used. For the QCT image of each subject, the ground truth for the proximal femur was delineated by a well-trained scientist. During the experiments for the entire cohort then for male and female subjects separately, 90% of the subjects were used in 10-fold cross-validation for training and internal validation, and to select the optimal parameters of the proposed models; the rest of the subjects were used to evaluate the performance of models. Results: Visual comparison demonstrated high agreement between the model prediction and ground truth contours of the proximal femur portion of the QCT images. In the entire cohort, the proposed model achieved a Dice score of 0.9815, a sensitivity of 0.9852 and a specificity of 0.9992. In addition, an R2 score of 0.9956 (p<0.001) was obtained when comparing the volumes measured by our model prediction with the ground truth. Conclusion: This method shows a great promise for clinical application to QCT and QCT-based finite element analysis of the proximal femur for evaluating osteoporosis and hip fracture risk.

研究动机与目标

• 开发一种基于深度学习的自动化方法，用于从QCT图像中分割近端股骨，以支持骨健康状况的临床评估。
• 克服基于QCT的骨密度和生物力学分析中手动分割耗时且主观的问题。
• 通过使用3D全卷积神经网络架构，提高分割的准确性和可重复性。
• 在包含397名QCT受试者的大型队列上验证该方法，包括对男性和女性亚组的独立分析。
• 实现在基于QCT的有限元分析中可靠应用，以评估骨质疏松症和髋部骨折风险。

提出的方法

• 采用一种3D U-Net变体，即V-Net，作为骨干网络，实现从QCT体积中端到端的全卷积3D分割。
• 该模型使用复合损失函数，结合Dice损失以处理类别不平衡问题，以及L2正则化以提高泛化能力并防止过拟合。
• 训练策略在数据集的90%（357名受试者）上采用10折交叉验证，用于超参数调优和模型选择。
• 剩余10%的受试者（40名）用于在未见数据上进行最终性能评估。
• 真实标签分割由经验丰富的专家创建，确保了高质量的标注以供监督。
• 推理过程完全自动化，预测结果直接输出体素级分割掩码。

实验结果

研究问题

• RQ1 基于深度学习的3D分割模型能否在自动分割QCT图像中的近端股骨时实现高精度和强鲁棒性？
• RQ2 所提出的复合损失函数（Dice + L2）与标准损失函数相比，在处理类别不平衡的医学图像数据时，能否显著提升分割性能？
• RQ3 在包含男性和女性的大型多样化QCT受试者队列中，该模型的分割性能如何？
• RQ4 该模型预测的体积与真实体积之间的相关性如何，表明其在下游生物力学分析中的可靠性？
• RQ5 该模型能否有效应用于临床QCT影像流程中，以实现骨质疏松症和髋部骨折风险评估？

主要发现

• 该方法在整体队列中达到0.9815的Dice分数，表明预测分割与真实分割之间近乎完美的重叠。
• 灵敏度为0.9852，表明模型能够正确识别近端股骨中绝大多数真正的阳性体素。
• 特异性达到0.9992，表明误报检测极少，这对准确估算骨体积至关重要。
• 与真实体积相比，模型预测的骨体积R²为0.9956（p < 0.001），证实了强烈的线性相关性。
• 视觉检查确认了预测结果与专家勾画轮廓之间具有高度的解剖学保真度和一致性。
• 该方法在男性和女性受试者中均表现出强泛化能力，在亚组分析中性能一致。