[论文解读] Automated lung segmentation from CT images of normal and COVID-19 pneumonia patients

Faeze Gholamiankhah, Samaneh Mostafapour+5 more|arXiv (Cornell University)|Apr 5, 2021

COVID-19 diagnosis using AI被引用 10

一句话总结

本研究提出了一种基于深度学习的残差神经网络，用于对正常及新型冠状病毒肺炎（COVID-19）肺炎患者的CT扫描图像实现肺部分割的自动化。该模型在1,200例经确诊的COVID-19病例上进行训练，其参考肺部掩膜由人工/半自动方法生成。模型在正常受试者中的Dice相似性系数（DSC）达到0.980，在COVID-19患者中达到0.971，表明即使在疾病引起的异常情况下，该模型仍具有高精度和强鲁棒性，能够实现肺部的准确分割。

ABSTRACT

Automated semantic image segmentation is an essential step in quantitative image analysis and disease diagnosis. This study investigates the performance of a deep learning-based model for lung segmentation from CT images for normal and COVID-19 patients. Chest CT images and corresponding lung masks of 1200 confirmed COVID-19 cases were used for training a residual neural network. The reference lung masks were generated through semi-automated/manual segmentation of the CT images. The performance of the model was evaluated on two distinct external test datasets including 120 normal and COVID-19 subjects, and the results of these groups were compared to each other. Different evaluation metrics such as dice coefficient (DSC), mean absolute error (MAE), relative mean HU difference, and relative volume difference were calculated to assess the accuracy of the predicted lung masks. The proposed deep learning method achieved DSC of 0.980 and 0.971 for normal and COVID-19 subjects, respectively, demonstrating significant overlap between predicted and reference lung masks. Moreover, MAEs of 0.037 HU and 0.061 HU, relative mean HU difference of -2.679% and -4.403%, and relative volume difference of 2.405% and 5.928% were obtained for normal and COVID-19 subjects, respectively. The comparable performance in lung segmentation of the normal and COVID-19 patients indicates the accuracy of the model for the identification of the lung tissue in the presence of the COVID-19 induced infections (though slightly better performance was observed for normal patients). The promising results achieved by the proposed deep learning-based model demonstrated its reliability in COVID-19 lung segmentation. This prerequisite step would lead to a more efficient and robust pneumonia lesion analysis.

研究动机与目标

• 开发一种适用于正常及新型冠状病毒肺炎（COVID-19）肺炎患者CT图像的可靠自动化肺部分割方法。
• 评估模型在包括因COVID-19导致严重肺部受累在内的多样化患者群体中的性能表现。
• 为肺部疾病中肺部病灶的定量分析建立稳健的预处理步骤。
• 利用多种定量指标比较正常肺部与COVID-19影响肺部分割的准确性。
• 支持后续应用，如疾病进展监测和治疗反应评估。

提出的方法

• 在1,200例经确诊的COVID-19 CT扫描图像上训练了残差神经网络（ResNet），并配有相应的参考肺部掩膜。
• 通过半自动和人工分割方法生成参考肺部掩膜，以确保高质量的真实标签。
• 模型在两个独立的外部测试数据集上进行评估：120例正常患者和120例COVID-19患者。
• 采用标准指标评估性能：Dice相似性系数（DSC）、平均绝对误差（MAE）、相对平均Hounsfield单位（HU）差异以及相对体积差异。
• 网络架构利用跳跃连接以改善梯度流动并提升深度学习中的训练稳定性。
• 应用数据增强和归一化技术，以增强模型在不同扫描强度和解剖变异下的泛化能力。

实验结果

研究问题

• RQ1 深度学习模型是否能在严重COVID-19肺炎患者CT扫描中实现高精度肺部分割？
• RQ2 模型在COVID-19影响肺部中的表现与在正常肺部中的表现相比如何？
• RQ3 COVID-19患者中肺部浸润和磨玻璃样改变的存在在多大程度上影响分割精度？
• RQ4 该模型是否具备足够的鲁棒性，可在不同人群特征和扫描协议下保持高性能？
• RQ5 自动化肺部分割能否作为肺炎肺部病灶后续定量分析的可靠基础？

主要发现

• 对于正常受试者，模型的Dice相似性系数（DSC）为0.980，表明与参考肺部掩膜的重叠近乎完美。
• 对于COVID-19患者，DSC为0.971，表明即使在广泛肺部受累的情况下，模型仍表现出强劲的分割性能。
• 正常受试者的平均绝对误差（MAE）为0.037 HU，COVID-19患者的MAE为0.061 HU，反映出高精度的HU值一致性。
• 正常受试者的相对平均HU差异为-2.679%，COVID-19患者的相对平均HU差异为-4.403%，表明HU值偏差较小。
• 正常受试者的相对体积差异为2.405%，COVID-19患者的相对体积差异为5.928%，表明体积估计精度可接受。
• 模型在正常肺部上的表现略优，但在COVID-19影响肺部中仍实现了高精度，证实了其鲁棒性。