Department of Respiration, Gong An County People's Hospital, Jingzhou, China.
Department of Laboratory Medicine, The Second Affiliated Hospital, Guangzhou University of Chinese Medicine, Guangzhou, China.
J Med Internet Res. 2021 Feb 22;23(2):e23390. doi: 10.2196/23390.
The initial symptoms of patients with COVID-19 are very much like those of patients with community-acquired pneumonia (CAP); it is difficult to distinguish COVID-19 from CAP with clinical symptoms and imaging examination.
The objective of our study was to construct an effective model for the early identification of COVID-19 that would also distinguish it from CAP.
The clinical laboratory indicators (CLIs) of 61 COVID-19 patients and 60 CAP patients were analyzed retrospectively. Random combinations of various CLIs (ie, CLI combinations) were utilized to establish COVID-19 versus CAP classifiers with machine learning algorithms, including random forest classifier (RFC), logistic regression classifier, and gradient boosting classifier (GBC). The performance of the classifiers was assessed by calculating the area under the receiver operating characteristic curve (AUROC) and recall rate in COVID-19 prediction using the test data set.
The classifiers that were constructed with three algorithms from 43 CLI combinations showed high performance (recall rate >0.9 and AUROC >0.85) in COVID-19 prediction for the test data set. Among the high-performance classifiers, several CLIs showed a high usage rate; these included procalcitonin (PCT), mean corpuscular hemoglobin concentration (MCHC), uric acid, albumin, albumin to globulin ratio (AGR), neutrophil count, red blood cell (RBC) count, monocyte count, basophil count, and white blood cell (WBC) count. They also had high feature importance except for basophil count. The feature combination (FC) of PCT, AGR, uric acid, WBC count, neutrophil count, basophil count, RBC count, and MCHC was the representative one among the nine FCs used to construct the classifiers with an AUROC equal to 1.0 when using the RFC or GBC algorithms. Replacing any CLI in these FCs would lead to a significant reduction in the performance of the classifiers that were built with them.
The classifiers constructed with only a few specific CLIs could efficiently distinguish COVID-19 from CAP, which could help clinicians perform early isolation and centralized management of COVID-19 patients.
COVID-19 患者的初始症状与社区获得性肺炎（CAP）患者非常相似；仅凭临床症状和影像学检查难以将 COVID-19 与 CAP 区分开来。
本研究旨在构建一种有效的 COVID-19 早期识别模型，同时将其与 CAP 区分开来。
回顾性分析 61 例 COVID-19 患者和 60 例 CAP 患者的临床实验室指标（CLIs）。利用机器学习算法（包括随机森林分类器（RFC）、逻辑回归分类器和梯度提升分类器（GBC））对各种 CLIs 的随机组合（即 CLI 组合）进行分析，建立 COVID-19 与 CAP 分类器。使用测试数据集评估分类器在 COVID-19 预测中的性能，通过计算受试者工作特征曲线（ROC）下面积（AUROC）和召回率。
3 种算法从 43 个 CLI 组合中构建的分类器在测试数据集的 COVID-19 预测中表现出较高的性能（召回率＞0.9，AUROC＞0.85）。在高性能分类器中，有几个 CLIs 的使用率较高，包括降钙素原（PCT）、平均红细胞血红蛋白浓度（MCHC）、尿酸、白蛋白、白蛋白与球蛋白比值（AGR）、中性粒细胞计数、红细胞计数（RBC）计数、单核细胞计数、嗜碱性粒细胞计数和白细胞计数（WBC）计数。除了嗜碱性粒细胞计数外，它们的特征重要性也很高。在使用 RFC 或 GBC 算法时，PCT、AGR、尿酸、WBC 计数、中性粒细胞计数、嗜碱性粒细胞计数、RBC 计数和 MCHC 的特征组合（FC）是构建分类器的 9 个 FC 中最具代表性的一个，AUROC 为 1.0。在这些 FC 中替换任何 CLIs 都会导致使用它们构建的分类器性能显著下降。
仅使用少数特定 CLIs 构建的分类器可有效地将 COVID-19 与 CAP 区分开来，这有助于临床医生对 COVID-19 患者进行早期隔离和集中管理。
