Diabetes Predictor

Link to Site DiabetesPredictor

Overview

Diabetes, a chronic and life-threatening disease, presents challenges in timely identification. Leveraging machine learning, our project develops a predictive model for early diabetes detection. Five ML models were evaluated on the 'Early stage diabetes risk prediction dataset,' resulting in Random Forest achieving 97.2% accuracy. Deployed via MySQL and Streamlite, our system offers a comprehensive web platform for user authentication, data storage, and interactive access to diabetes-related information. Our solution aims to enhance early intervention, reducing diabetes-related complications.

Usage

1. First download the zip file nad extract the file then follow below instructions according to your requirement.
2. TO DEPLOY THE APP USING STREAMLIT AND GITHUB
   • Go to TO DEPLOY folder
   • Directly upload the folder diabetespredictor-main to Github
   • Create a new app in Streamlit Community Cloud
   • Run the app
3. IF YOU WANT TO RUN THE APP ON LOCAL HOST
   • Go to PROJECT folder
   • Open the folder DIABETESPREDICTOR in vscode
   • Run mainpage.py file

Results

Prediction Page

Diabetes Negative Prediction

Diabetes Positive Prediction

User Reports

If user wants to Download only Particular Reports he/she can select those report ID’s. They can just click on Generate to download all reports.

Generated PDF