Diabetes Risk Predictor 🩺

A simple machine learning app that predicts the probability of diabetes from patient health data using Logistic Regression. Built with Python, scikit-learn, and Streamlit for an interactive and user-friendly interface.

Features

• Predicts diabetes probability based on key health metrics, including gender, age, hypertension, heart disease, smoking history, BMI, HbA1c and blood glucose levels.
• Displays prediction (risk factor) in a easy-to-read format
• Interactive web app interface

Demo

Follow the link: https://diabetes-predictor-model-app.streamlit.app!

How to Run Locally

1. Clone the repository:

git clone 
cd diabetes-risk-predictor

2. Install dependencies:

pip install -r requirements.txt

3. Run the Streamlit app:

streamlit run app.py

4. Open the app in your browser (Streamlit will provide the link).

Dataset

This model was trained on a publicly available diabetes dataset from Kaggle (link: https://www.kaggle.com/datasets/iammustafatz/diabetes-prediction-dataset?resource=download).

Technologies Used

• Python
• scikit-learn (Logistic Regression, StandardScaler)
• Streamlit
• Pandas
• Joblib