Effective Diagnosis of Diabetes Mellitus Using Neural Networks and its Hardware Implementation on FPGA

• Published in: International journal of computer science and information security Vol. 15; no. 1; p. 519
• Main Author: Ananthalakshmi, A V
• Format: Journal Article
• Language: English
• Published: Pittsburgh L J S Publishing 01.01.2017
• Subjects: Blood; Classification; Diabetes; Diabetes mellitus; Diagnosis; Fasting; Glucose; Insulin; Mathematical models; Matlab; Neural networks; Patients; Regulatory approval; Sugars
• ISSN: 1947-5500

Diabetes Mellitus (DM), commonly referred to as Diabetes, is a group of metabolic diseases in which there are high blood sugar levels over a prolonged period. Thus, the objective of this work is to predict the chances of diabetes in a person and if a person has diabetes then the level of insulin dosage required for the patient is prescribed. Parameters such as Fasting Blood Sugar levels (FBS), Postprandial Plasma Glucose Test (PPBS), Urea, Creatinine and Haemoglobin factors are considered in order to detect the diabetes in a person. Using neural network feed forward prediction model in conjunction with back propagation algorithm, and based on the given training data set, it is predicted whether a subject has diabetes or not using Matlab. Finally, the hardware implementation of the Diagnosis of Diabetes Mellitus based on Neural network concept is carried out in Xilinx Spartan6 FPGA XC6SLX9. Functionality of the design is verified using Verilog HDL module.