Skip to content

A predictive model for assessing diabetes risk using health indicators such as BMI, glucose levels, and insulin. Includes a real-time Shiny app for user-friendly risk assessments. Achieved an AUC of 0.8349 using logistic regression and recursive feature elimination.

Notifications You must be signed in to change notification settings

iRubeel/Diabetes-Risk-Prediction

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 

Repository files navigation

Diabetes Risk Prediction

This project involves developing a predictive model to assess the risk of diabetes using health indicators like BMI, glucose levels, and blood pressure. A real-time risk assessment app was built using Shiny, allowing users to input their health data and get diabetes risk predictions.

Table of Contents

Project Motivation

This project was inspired by a family member who developed diabetes. The goal was to create a tool that could provide a quick assessment of diabetes risk using logistic regression and Recursive Feature Elimination (RFE) based on user inputs like BMI, blood pressure, and glucose levels.

Technologies Used

  • Python (for data preprocessing, model development)
  • Logistic Regression (predictive model)
  • Recursive Feature Elimination (RFE) (feature selection)
  • Shiny (R) (for web application)
  • Pandas (data manipulation)
  • scikit-learn (model training and evaluation)
  • Matplotlib (visualizations)
  • Shiny (for developing the web interface)

Data Source

The dataset was sourced from UCI Machine Learning Repository and consists of various health metrics related to diabetes risk.

Methodology

  1. Data Preprocessing: Cleaning and scaling the dataset for model input.
  2. Feature Selection: Recursive Feature Elimination (RFE) was used to identify the most important health indicators.
  3. Modeling: A Logistic Regression model was trained on the processed data.
  4. Evaluation: The model's performance was evaluated using various metrics, including the AUC (Area Under Curve), achieving an AUC of 0.8349.
  5. Deployment: A Shiny app was created to allow real-time diabetes risk prediction based on user input.

Results

The Logistic Regression model achieved an AUC of 0.8349, indicating good performance. The Shiny application allows users to interactively assess their diabetes risk.

Shiny App

The Shiny app allows users to input health metrics such as BMI, blood pressure, and glucose levels. The app outputs a diabetes risk score in real-time based on the trained model.

About

A predictive model for assessing diabetes risk using health indicators such as BMI, glucose levels, and insulin. Includes a real-time Shiny app for user-friendly risk assessments. Achieved an AUC of 0.8349 using logistic regression and recursive feature elimination.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages