Prediction of type 2 diabetes mellitus using hematological factors based on machine learning approaches: A cohort study analysis

Background: Type 2 Diabetes mellitus (T2DM) is a significant public health problem globally. The diagnosis and management of diabetes are critical to reduce the complications of diabetes including cardiovascular disease and cancer. This study was designed to assess the potential association between T2DM and several routinely measured hematological parameters. Method: This study was a subsample of 9000 adults aged 35–65 years recruited as part of Mashhad stroke and heart atherosclerotic disorder (MASHAD) cohort study population. Data mining techniques including logistic regression (LR), decision tree (DT) and bootstrap forest (BF) algorithms were applied to analyze data. All data analyses were performed using SPSS version 22 and SAS JMP Pro version 13 at a significant level of 0.05. Results: Based on the performance indices, the BF model gave high accuracy, precision, specificity, and AUC compared to the other models. Thus, according to all the assessed hematological factors, the most effective risk factors for predicting the development of T2DM in the BF model were age and WBC. Conclusion: In summary, the BF model represented a better performance to predict T2DM. Also, our selected model provides valuable information on critical determinants to predict T2DM like age and WBC.