Use of Machine Learning and Routine Laboratory Tests for Diabetes Mellitus Screening

Figure 1

A proposed pipeline for studying the HbA1c classification. There are four main steps: data collection, data preprocessing, model training, and performance evaluation. Three datasets were created: HPD (H = healthy; P = prediabetes; D = diabetes), HN (H = healthy; N = no healthy []), and ND (N = no diabetes []; D = diabetes). Datasets with the suffix refer to the regression models with later classification.