MODEL BASED ON SOFT SETS AND APPROXIMATE SET METHODS FOR THE DIAGNOSIS OF PERIPHERAL NEUROPATHY DIABETIC

Abstract

Diabetic neuropathy is the most frequent chronic microangiopathic complication that generates greater disability due to amputations and mortality in diabetes mellitus. Hence, efforts are considered necessary to contribute with essential methodological recommendations for students of medical sciences, patients and specialists, in the management of diabetic peripheral neuropathy (DPN) based on the best scientific evidence. In this disease, the clinical specialty is crucial for a correct diagnosis, however, uncertainty is appreciated in the diagnosis of this disease. The aim of this article is to propose a soft set-based model for the diagnosis of diabetic peripheral neuropathy. Soft sets are another way of modeling uncertainty, they also generalize fuzzy sets, with the advantage that they do not necessarily need of membership functions to be defined. The model uses a hybrid between soft sets and methods of rough sets. The latter ones are also used to model uncertainty and are useful for arriving at knowledge-related rules from data.