Early Prediction Model for Type-2 Diabetes Based on Lifestyle

¹ Department of Information Technology, Ramrao Adik Institute of Technology, Navi Mumbai, India
² Department of Information Technology, Ramrao Adik Institute of Technology, Navi Mumbai, India
³ Department of Information Technology, Ramrao Adik Institute of Technology, Navi Mumbai, India
⁴ Department of Information Technology, Ramrao Adik Institute of Technology, Navi Mumbai, India

^* e-mail: e-mail: kirti50hirnak@gmail.com
^** e-mail: nikitachaudhari863@gmail.com
^*** e-mail: akshaysingh.mustang@gmail.com
^**** e-mail: deepatli.patil@rait.ac.in

Abstract

Nowadays diabetes has become a chronic disease that may cause many complications. There are some symptoms of diabetes such as increased appetite, blurry vision, and extreme fatigue, etc. As the increasing deformity in present years the number of diabetic patients from the whole world will reach to 642 million. Diabetes accuracy is very difficult to know so in order to cure this disease. These causes us to concentrate more there to make some changes that will reduce these numbers. So to minimize these numbers of diabetes, we researched various algorithms and methods. The proposed method focuses on extracting the attributes that gives a result in early detection of Diabetes Mellitus in patients. Various existing processes provide just a result as the patient has diabetes or not which will require the patients to visit a diagnostic centers or to a doctor. So we proposed a system based on deep learning approaches that will help to solve a serious problem. These systems take collaborative inputs from dataset to give prediction with random forest algorithm which gives more accurate results.