A Novel Fuzzy System For Predicting Drug Doses In Hypertensive Patients With Chronic Kidney Disease And Type 2 Diabetes Mellitus

Kadime Gogebakan Perez

Istanbul Technical University, Turkey

Biography :

Kadime Gogebakan, completed her doctorate at Near East Uni­versity at the age of 45. She is an Information Technologies Direc­torate Engineer at the Northern Cyprus campus of Istanbul Tech­nical University, a leading univer­sity.

Abstract :

Incorrect drug dosages can lead to side effects, complica­tions, and even fatal outcomes for patients. Today, patients with multiple chronic diseases often receive prescriptions from different specialist physicians meanwhile and this in­creases the risk of dosage errors if proper care is not taken. In this work, we particularly focus on hyperintensive patients with Type-2 Diabetes Mellitus (T2DM) and Chronic Kidney Disease (CKD) since these two diseases are common and re­lated to each other. We proposed a novel fuzzy rule-based system that uses Glomerular Filtration Rate (GFR) values of patients to accurately model CKD stages of patients with uncertainty. In the proposed system, first we received ex­pert opinion from three independent physicians regrading to drugs and drug doses in hyperintensive patients depending on their CKD stages; (a) 18 drugs for T2DM treatment and (b) 66 drugs for CKD treatment. Subsequently, depending on fuzzified input GFR value of patients and fuzzy rules are de­signed based on the expert knowledge on drug doses. The proposed system, generates a total of 504 drug dose val­ues for both diseases. Finally, drug doses for invidual patients are estimated for the treatment of both T2DM and CKD. To the best of our knowledge, none of the previous works use a fuzzy system to arrange drug doses for hyperintensive pa­tients with T2DM and CKD, and this is the contribution of the proposed work.