The Detection of Prameha (Diabetes) in Ayurvedic Way with the Help of Fuzzy Deep Learning

摘要

A society's awareness about its fitness, determines betterment of that society's life style. Diabetes is a disease which is born because of unhealthy life style. This motivates our work in area of Diabetes detection. In the recent time period the use of Convolutional Neural Network (CNN) to design the diabetes detection framework is getting attention since it is outperforming most of the application areas involving prediction. Here is an attempt to apply Deep Convolutional Neural Network on Ayurvedic data set of Indian population. Samples were collected from a private hospital without privacy violating attributes. The objective of this research is to find the applicability of CNN in diabetes detection. Additionally, it uses fuzzification of data so that data could be used with Convolutional Neural Network. Results show that proposed hybrid approach of integrating fuzzification with CNN has outperformed classical Neural Networks and other state of the art algorithms.