Differentiating novel coronavirus pneumonia from general pneumonia based on machine learning
Abstract Background Chest CT screening as supplementary means is crucial in diagnosing novel coronavirus pneumonia (COVID-19) with high sensitivity and popularity. Machine learning was adept in discovering intricate structures from CT images and achieved expert-level performance in medical image analysis. Methods To develop and validate an integrated machine learning framework on chest CT images for differentiating COVID-19 from common pneumonia (CP). Seventy-three confirmed COVID-19 cases were consecutively enrolled together with twenty-seven confirmed common pneumonia patients from Ruian People’s Hospital, from January 2020 to March 2020. Statistical textual features of COVID-19 and CP images were extracted. After feature selection, the reserved features were applied to the ensemble of bagged tree (EBT) and four other machine learning classifiers with 10-fold cross-validation. Results The classification accuracy, precision, sensitivity, specificity and F1 score of our proposed method are 91.66%, 97.91%, 85.26%, 98.15% and 91.15% respectively. The AUC of its receiver operating characteristic is 0.98. Conclusions The experimental results indicate that the EBT algorithm with statistical textural features on chest CT for differentiating COVID-19 from common pneumonia achieved high transferability, efficiency, specificity, and impressive accuracy.