Objective: To identify the potential biomarkers for predicting depression in diabetes mellitus using support vector machine to analyze routine biochemical tests and vital signs between two groups: subjects with both diabetes mellitus and depression, and subjects with diabetes mellitus alone. Methods: Electronic medical records upon admission and biochemical tests and vital signs of 135 patients with both diabetes mellitus and depression and 187 patients with diabetes mellitus alone were identified for this retrospective study. After matching on factors of age and sex, the two groups (n = 72 for each group) were classified by the recursive feature elimination-based support vector machine, of which, the training data, validation data, and testing data were split for ranking the parameters, determine the optimal parameters, and assess classification performance. The biomarkers were identified by 10-fold cross validation. Results: The experimental results identified 8 predictive biomarkers with classification accuracy of 78%. The 8 biomarkers are magnesium, cholesterol, AST/ALT, percentage of monocytes, bilirubin indirect, triglyceride, lactic dehydrogenase, and diastolic blood pressure. Receiver operating characteristic curve analysis was also adopted with area under the curve being 0.72. Conclusions: Some biochemical parameters may be potential biomarkers to predict depression among the subjects with diabetes mellitus.
Keywords: biomarkers; depression; diabetes mellitus; machine learning method; support vector machine.
Copyright © 2021 Song, Zheng, Zhang, Wang, Deng, Wang, Guo, Li and Ma.