Almost 25% of COVID-19 patients end up in ICU needing critical mechanical ventilation support. There is currently no validated objective way to predict which patients will end up needing ventilator support, when the disease is mild and not progressed. N = 869 patients from two sites (D ₁ : N = 822, D ₂ : N = 47) with baseline clinical characteristics and chest CT scans were considered for this study. The entire dataset was randomly divided into 70% training, D ₁ ^train (N = 606) and 30% test-set (D ^test : D ₁ ^test (N = 216) + D ₂ (N = 47)). An expert radiologist delineated ground-glass-opacities (GGOs) and consolidation regions on a subset of D ₁ ^train , (D ₁ ^train_sub , N = 88). These regions were automatically segmented and used along with their corresponding CT volumes to train an imaging AI predictor (AIP) on D ₁ ^train to predict the need of mechanical ventilators for COVID-19 patients. Finally, top five prognostic clinical factors selected using univariate analysis were integrated with AIP to construct an integrated clinical and AI imaging nomogram (ClAIN). Univariate analysis identified lactate dehydrogenase, prothrombin time, aspartate aminotransferase, %lymphocytes, albumin as top five prognostic clinical features. AIP yielded an AUC of 0.81 on D ^test and was independently prognostic irrespective of other clinical parameters on multivariable analysis (p<0.001). ClAIN improved the performance over AIP yielding an AUC of 0.84 (p = 0.04) on D ^test . ClAIN outperformed AIP in predicting which COVID-19 patients ended up needing a ventilator. Our results across multiple sites suggest that ClAIN could help identify COVID-19 with severe disease more precisely and likely to end up on a life-saving mechanical ventilation.

中文翻译：

---

用于预测 COVID-19 患者严重程度和呼吸机支持需求的综合临床和基于 CT 的人工智能列线图：一项多站点研究

近 25% 的 COVID-19 患者最终进入 ICU，需要关键的机械通气支持。目前还没有经过验证的客观方法来预测哪些患者最终需要呼吸机支持，当疾病是轻微的并且没有进展时。本研究考虑了来自两个地点的 N = 869 名患者（D ₁ ：N = 822，D ₂ ：N = 47），具有基线临床特征和胸部 CT 扫描。整个数据集随机分为 70% 训练、D ₁ ^训练(N = 606) 和 30% 测试集 (D ^测试 ：D ₁ ^测试(N = 216) + D ₂ (N = 47))。_{一位放射科专家在 D 1}的子集上描绘了磨玻璃影 (GGO) 和实变区域 ^火车 , (D ₁ ^train_sub , N = 88)。这些区域被自动分割并与它们相应的 CT 体积一起用于训练 D ₁ ^列车上的成像 AI 预测器 (AIP)，以预测 COVID-19 患者对机械呼吸机的需求。最后，将使用单变量分析选择的前五个预后临床因素与 AIP 相结合，以构建综合临床和 AI 成像列线图 (ClAIN)。单变量分析将乳酸脱氢酶、凝血酶原时间、天冬氨酸氨基转移酶、淋巴细胞百分比、白蛋白确定为前五个预后临床特征。^{AIP 在 D测试}中产生了 0.81 的 AUC 无论多变量分析中的其他临床参数如何，都具有独立的预后意义（p<0.001）。ClAIN 提高了 AIP 的性能，在 D ^测试中产生了 0.84 (p = 0.04) 的 AUC 。在预测哪些 COVID-19 患者最终需要呼吸机方面，ClAIN 的表现优于 AIP。我们在多个站点的结果表明，ClAIN 可以帮助更准确地识别患有严重疾病的 COVID-19，并且可能最终采用挽救生命的机械通气。