Computed Tomography Images under the Nomogram Mathematical Prediction Model in the Treatment of Cerebral Infarction Complicated with Nonvalvular Atrial Fibrillation and the Impacts of Virus Infection

Abstract

The aim of this work was to explore the effect of the nomogram mathematical model on the treatment of cerebral infarction complicated with nonvalvular atrial fibrillation (NVAF) and viral infection. The data were scanned by a circular trajectory fan beam isometric scanning mode system (scanning system), and the speckle noise of computed tomography (CT) images was smoothed by Lee filtering. 52 patients with postoperative recurrent viral infection (RVI group) and 248 patients without postoperative recurrent viral infection (NRVI group) were selected for retrospective analysis. The mathematical model curve was then analyzed through calibration plots and decision curves to predict the accuracy of the mathematical model. The results showed that the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy based on the training set were 0.7868, 0.7634, 0.6982, and 0.7146, respectively. The AUC, sensitivity, specificity, and accuracy based on the validation set were 0.7623, 0.7734, 0.6882, and 0.6948, respectively. There was no significant difference in the AUC between the two groups (P > 0.05), indicating that the nomogram mathematical prediction model had high repeatability. In conclusion, CT images based on the nomogram mathematical prediction model had good predictive ability in the treatment of cerebral infarction complicated with NVAF.