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

Figure 2

Datasets used with the models. (a) The three subdatasets used with the classification models: healthy, prediabetes, and diabetes (HPD); healthy and no healthy (HN); and no diabetes and diabetes (ND). (b) The three subdatasets used with the regression models (i.e., classification after regression): healthy, prediabetes, and diabetes (HPDr); healthy and no healthy (HNr), being no healthy formed by prediabetes and diabetes; and no diabetes and diabetes (NDr), being no diabetes formed by healthy and prediabetes.

(a)

(b)