[论文解读] Development and Validation of a Deep Learning Model for Prediction of Severe Outcomes in Suspected COVID-19 Infection

Buch, Aoxiao Zhong+9 more|arXiv (Cornell University)|Mar 21, 2021

COVID-19 diagnosis using AI参考文献 20被引用 4

一句话总结

本研究开发并验证了一种深度学习模型 CO-RISK，该模型整合电子健康记录（EHR）数据与胸部X光（CXR）影像，以预测急诊科（ED）就诊后24–72小时内疑似COVID-19患者的严重结局。该模型在24小时时AUC达到0.95，73小时时为0.92，优于临床风险评分和医生判断，在ICU/病房分诊决策中表现更优。

ABSTRACT

COVID-19 patient triaging with predictive outcome of the patients upon first present to emergency department (ED) is crucial for improving patient prognosis, as well as better hospital resources management and cross-infection control. We trained a deep feature fusion model to predict patient outcomes, where the model inputs were EHR data including demographic information, co-morbidities, vital signs and laboratory measurements, plus patient's CXR images. The model output was patient outcomes defined as the most insensitive oxygen therapy required. For patients without CXR images, we employed Random Forest method for the prediction. Predictive risk scores for COVID-19 severe outcomes ("CO-RISK" score) were derived from model output and evaluated on the testing dataset, as well as compared to human performance. The study's dataset (the "MGB COVID Cohort") was constructed from all patients presenting to the Mass General Brigham (MGB) healthcare system from March 1st to June 1st, 2020. ED visits with incomplete or erroneous data were excluded. Patients with no test order for COVID or confirmed negative test results were excluded. Patients under the age of 15 were also excluded. Finally, electronic health record (EHR) data from a total of 11060 COVID-19 confirmed or suspected patients were used in this study. Chest X-ray (CXR) images were also collected from each patient if available. Results show that CO-RISK score achieved area under the Curve (AUC) of predicting MV/death (i.e. severe outcomes) in 24 hours of 0.95, and 0.92 in 72 hours on the testing dataset. The model shows superior performance to the commonly used risk scores in ED (CURB-65 and MEWS). Comparing with physician's decisions, CO-RISK score has demonstrated superior performance to human in making ICU/floor decisions.

研究动机与目标

• 通过预测需要机械通气或死亡等严重结局，改进急诊科对疑似COVID-19患者的早期分诊。
• 将多模态数据——EHR（人口统计学、共病、生命体征、检验结果）与CXR影像——整合到统一的预测模型中。
• 开发一种临床可操作的风险评分（CO-RISK），以支持医疗环境中的资源调配与感染控制。
• 将模型性能与既有的临床风险评分（CURB-65、MEWS）及医生判断进行验证。
• 通过使用随机森林替代方案，评估在无可用CXR影像患者中的模型稳健性。

提出的方法

• 基于EHR与CXR数据训练了一个深度特征融合模型，以预测最需要的氧疗强度作为严重结局的代理指标。
• 对于无CXR影像的患者，采用随机森林模型作为备用方案，以维持预测能力。
• 模型在马萨诸塞综合医院布兰德医疗系统（Mass General Brigham healthcare system）的11,060例确诊或疑似COVID-19患者队列中进行训练（2020年3月1日至6月1日）。
• CXR影像通过卷积神经网络（CNN）架构处理，以提取影像学特征，并与表格型EHR特征融合。
• 最终的CO-RISK评分由模型输出推导得出，并经过校准以增强临床可解释性与风险分层能力。
• 通过在保留测试集上计算受试者工作特征曲线下面积（AUC）评估模型性能。

实验结果

研究问题

• RQ1 是否能够开发一种融合EHR与CXR数据的深度学习模型，以比现有临床风险评分更准确地预测严重COVID-19结局（如机械通气或死亡）？
• RQ2 CO-RISK模型在急诊科医生做出ICU或住院分诊决策时的表现如何？
• RQ3 当CXR影像不可用时，模型的预测性能如何？备用随机森林模型的表现如何？
• RQ4 与仅使用EHR的模型相比，影像数据的引入是否显著提升了预测准确性？
• RQ5 CO-RISK评分是否可被可靠地用于早期分诊，以优化医院资源调配与感染控制？

主要发现

• 在测试数据集中，CO-RISK模型在急诊科就诊后24小时内预测严重结局（机械通气/死亡）的AUC达到0.95。
• 在72小时时，模型仍保持优异性能，AUC为0.92，表明其预测准确性具有时间上的持续性。
• 与CURB-65和MEWS临床风险评分相比，CO-RISK在预测严重结局方面表现显著更优。
• 在区分需要ICU收治与普通病房治疗的患者方面，CO-RISK的表现优于医生的分诊决策。
• 即使在无可用CXR影像的患者中，该模型仍保持高性能，得益于随机森林备用方法。
• 本研究证实，EHR与影像数据的多模态整合可显著提升对严重COVID-19结局的早期预测能力。