Diagnosis Support for Diabetes with Ant Colony Optimization

Diabetes is one of the most serious health problems in humans. Both industrialized and developing nations struggle with the health effects of this disease, which is becoming more widespread. One of the most important characteristics of diabetes is that approximately half of diabetic people have inherited traits. Inadequate insulin production and poor pancreatic function are also major reasons for diabetes. Recently, genome-wide association studies have meaningful numerous additional genes that are probably associated with diabetes-like disease. However, the heritability of type 2 diabetes is largely unaccounted for by the widespread genetic variants that have been discovered thus far. A further component of this heritability may be explained by the interaction of two or more gene variants; however, complete interaction analyses are currently impossible or extremely computationally challenging. In the present study, ant colony optimization (ACO) algorithm methods were proposed for the diagnosis of diabetes. ACO has been used to successfully extract rule-based categorization systems in the field of data mining. The study summarizes the use of ACO to derive a set of rules for diabetes disease diagnosis.