Decision Tree as a screening tool for the diagnosis of the Metabolic Syndrome without doing a blood test

Metabolic Syndrome (MetS) is a pathology with a high probability of triggering the onset of diabetes, coronary heart disease, and other diseases. So, the more the data is analyzed, the closer it will be to finding a way to prevent or delay the occurrence of its atrocious results. Thus, the medical community is doing prevalence studies in several populations to analyze it. However, the cost of doing a blood test increases the budget. This article presents a screening tool to diagnose the MetS without doing a blood test using a decision tree model with the data of 615 subjects of a study conducted in Colombia anthropometric. We created a new decision tree and used the random subsampling technique to validate the proposed model to compare with other decision trees found in the art of state process. We proposed an excellent decision tree with the best Area under Receiver Operating Characteristic (AROC) of 82,58% that uses sex, waist perimeter, hip perimeter, systolic pressure, and diastolic pressure variables to predict the Metabolic Syndrome's diagnosis with the Harmonized Metabolic Syndrome. standard. Therefore, we invite using this model as a screening tool to diagnose the MetS early in a medical consultation or prevalence study.