Development of a fuzzy expert system for the nephropathy control assessment in patients with type 2 diabetes mellitus. (15th April 2017)

• Record Type: Journal Article
• Authors: Meza-Palacios, Ramiro; Aguilar-Lasserre, Alberto A.; Ureña-Bogarín, Enrique L.; Vázquez-Rodríguez, Carlos F.; Posada-Gómez, Rubén; Trujillo-Mata, Armín
• Is Part Of: Expert systems with applications. Volume 72(2017)
• Page Start: 335
• Page End: 343
• Publication Date: 2017-04-15
• Subjects: Fuzzy expert system -- Response surface -- Expertise -- Nephropathy control -- Type 2 diabetes mellitus; ANN Artificial neural networks -- CIs confidence intervals -- CPG clinical practice guidelines -- ES Expert system (s) -- ESRD End-stage renal disease -- FES Fuzzy expert system -- FL Fuzzy logic -- GFR Glomerular filtration rate -- T2DM Type 2 diabetes mellitus -- T1DM Type 1 diabetes mellitus -- T2DMA Type 2 diabetes mellitus age; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2016.10.053
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: A fuzzy expert system to assess the nephropathy control is proposed. The success rate of the fuzzy expert system is 93.33%. The use of this system allows reduce the lack of control and wrong treatments. The system would be useful if it is adapted to the diagnosis of other pathologies. Recognition of predictor variables of lack of control avoid irreversible damage in kidneys. Abstract: Diabetic nephropathy is a life-threatening complication if not controlled properly. Early detection and effective control prevent its progression. In this study, the development of a Fuzzy Expert System (FES) is proposed to help doctors assess the nephropathy control in patients with Type 2 Diabetes Mellitus (T2DM). The study is based on a FES that was developed with the use of Clinical Practice Guidelines (CPG), data bases and the expertise of a team of doctors. It considers the use of input variables such as Glomerular Filtration Rate (GFR), serum creatinine, blood glucose, Type 2 Diabetes Mellitus Age (T2DMA), uric acid, hypertension and dyslipidemia. All these factors, give an efficient nephropathy control assessment. Sixty tests were performed using the expertise of a team of doctors, the expected results were compared with those estimated by the FES (using the same cases), and it was observed that the FES succeeds in up to 93.33% of the cases. The response surface analysis shows that GFR, serum creatinine and hypertension have a greater impact in the nephropathy control. This system supports the doctors in nephropathy control, but it does not estimate the renal failure stages. Nephropathy control is a clinical problem that includes uncertainty and inaccuracy, so the use of a FES is recommended to overcome this problem, since fuzzy systems help to assess the inherent uncertainty degree.