Deep learning-based lesion subtyping and prediction of clinical outcomes in COVID-19 pneumonia using chest CT.
Deep learning-based lesion subtyping and prediction of clinical outcomes in COVID-19 pneumonia using chest CT.
复制标题
DOI：
10.1038/s41598-022-13298-8
复制
发表时间：
2022-06-07
影响因子：
4.6
通讯作者：
中科院分区：
文献类型：
--
作者：
这项工作的主要目标是开发和评估一种基于深度学习的人工智能系统，该系统能够自动识别、量化和表征COVID-19肺炎模式，以评估疾病严重程度并预测临床结果，并将预测性能与人类读者严重程度评估和全肺放射组学进行比较。我们提出了一种基于深度学习的方案，用于在非增强CT扫描中自动分割不同的病变亚型。自动病变量化用于预测临床结局。该技术已在2020年3月至7月期间回顾性收集的103名患者的多中心队列中进行了独立测试。使用重叠（Dice）和基于距离（Hausdorff和平均表面）的度量评估病变亚型的分割，而使用曲线下面积（AUC）评估预测临床相关结局的拟议系统。此外，还估计了其他指标，包括敏感性、特异性、阳性预测值和阴性预测值。正确计算了95%置信区间。脑实质损伤的自动估计值（%）与放射科医师的严重程度评分之间的一致性很强，斯皮尔曼相关系数（R）为0.83。病变亚型的自动量化能够预测患者死亡率、入住重症监护室（ICU）和机械通气需求，AUC分别为0.87、0.73和0.68。与放射科医生的解释和全肺放射组学相比，所提出的人工智能系统能够更好地预测那些临床相关的结果。总之，通过胸部CT平扫对COVID-19肺炎进行深度学习病变亚型分型，可以定量评估疾病严重程度，并根据全肺放射组学或放射科医生的严重程度评分更好地预测临床结局。
The main objective of this work is to develop and evaluate an artificial intelligence system based on deep learning capable of automatically identifying, quantifying, and characterizing COVID-19 pneumonia patterns in order to assess disease severity and predict clinical outcomes, and to compare the prediction performance with respect to human reader severity assessment and whole lung radiomics. We propose a deep learning based scheme to automatically segment the different lesion subtypes in nonenhanced CT scans. The automatic lesion quantification was used to predict clinical outcomes. The proposed technique has been independently tested in a multicentric cohort of 103 patients, retrospectively collected between March and July of 2020. Segmentation of lesion subtypes was evaluated using both overlapping (Dice) and distance-based (Hausdorff and average surface) metrics, while the proposed system to predict clinically relevant outcomes was assessed using the area under the curve (AUC). Additionally, other metrics including sensitivity, specificity, positive predictive value and negative predictive value were estimated. 95% confidence intervals were properly calculated. The agreement between the automatic estimate of parenchymal damage (%) and the radiologists’ severity scoring was strong, with a Spearman correlation coefficient (R) of 0.83. The automatic quantification of lesion subtypes was able to predict patient mortality, admission to the Intensive Care Units (ICU) and need for mechanical ventilation with an AUC of 0.87, 0.73 and 0.68 respectively. The proposed artificial intelligence system enabled a better prediction of those clinically relevant outcomes when compared to the radiologists’ interpretation and to whole lung radiomics. In conclusion, deep learning lesion subtyping in COVID-19 pneumonia from noncontrast chest CT enables quantitative assessment of disease severity and better prediction of clinical outcomes with respect to whole lung radiomics or radiologists’ severity score.
登录
查看更多内容
影响因子：
4.6
作者：
Guan X;Yao L;Tan Y;Shen Z;Zheng H;Zhou H;Gao Y;Li Y;Ji W;Zhang H;Wang J;Zhang M;Xu X
通讯作者：
Xu X
影响因子：
5.9
作者：
Feng Z;Shen H;Gao K;Su J;Yao S;Liu Q;Yan Z;Duan J;Yi D;Zhao H;Li H;Yu Q;Zhou W;Mao X;Ouyang X;Mei J;Zeng Q;Williams L;Ma X;Rong P;Hu D;Wang W
通讯作者：
Wang W
影响因子：
4.8
作者：
Harmouche, Rola;Ross, James C.;Estepar, Raul San Jose
通讯作者：
Estepar, Raul San Jose
影响因子：
4.8
作者：
Gietema, Hester A.;Mueller, Nestor L.;Coxson, Harvey O.
通讯作者：
Coxson, Harvey O.
影响因子：
19.7
作者：
Li, Lin;Qin, Lixin;Xia, Jun
通讯作者：
Xia, Jun