Objective To evaluate the acute exacerbation of chronic obstructive pulmonary disease(COPD)(AECOPD)status via combining clinical data,lung function parameters with CT radiomic features based on machine learning method.Methods A total of 343 COPD patients,including 158 AECOPD patients and 185 non-AECOPD patients were retrospectively selected and randomly divided into training and testing sets at a ratio of 7∶3.The radiomics features were calculated after automatically delineating the whole lung volume of interest(VOI).Five machine learning methods were used to construct the AECOPD diagnostic model,then the corresponding Radiomics score(Rad-score)was calculated in the training set and was validated in the testing set.The logistic-combined model was established after integrating age,Global Initiative for Chronic Obstructive Lung Disease(GOLD)classification,vital capacity(VC),forced vital capacity(FVC),forced expiratory volume in one second(FEV1),FEV1％pred,FEV1/FVC％,peak expiratory flow(PEF),maximum ventilatory volume(MVV),and Rad-score value.The area under the curve(AUC)of receiver operating characteristic(ROC)curve was calculated to evaluate the evaluated performance of all models.Results The logistic regression model had the best diagnostic performance,with AUC of 0.724 and 0.758 in the training and testing sets,respectively.The performance of the logistic-combined model to diagnose AECOPD was superior to that of the single logistic regression model,with the AUC of 0.777 and 0.760 in the training and testing sets,respectively.Conclusion A combination strategy including clinical data,lung function parameters,and CT radiomics may be helpful to diagnose AECOPD status,with moderate diagnostic performance.