🏥 HealthGuard AI 🚀 Inspiration

The idea for HealthGuard AI was driven by the growing need for early detection of chronic diseases, especially diabetes, which affects millions globally. Many people lack access to timely healthcare insights, and often diagnosis happens too late.

We wanted to build a system that leverages Artificial Intelligence to provide instant, accessible, and preventive health insights. The goal was simple:

“Empower individuals to understand their health risks before it’s too late.”

⚙️ What it does

HealthGuard AI is an AI-powered health risk prediction platform that:

Predicts the likelihood of diabetes using machine learning Provides a user-friendly dashboard with visual insights Allows users to: Sign up & log in securely Enter health parameters Get instant predictions Stores user history and feedback Displays analytics using charts and graphs

Mathematically, the model learns a function:

𝑓 ( 𝑥

)

𝑦 f(x)=y

Where:

𝑥 x = health parameters (glucose, BMI, age, etc.) 𝑦 y = predicted outcome (0 = Low Risk, 1 = High Risk) 🛠️ How we built it 🔹 Tech Stack Frontend: Streamlit Backend: Python Machine Learning: Scikit-learn (Random Forest) Database: SQLite → upgraded to PostgreSQL Visualization: Matplotlib, Seaborn 🔹 Development Process (SDLC) Planning Defined problem: early disease prediction Identified features: login, prediction, dashboard Design Designed UI using Streamlit Structured database schema (users, history, feedback) Development Built ML model using Random Forest Developed authentication system Integrated database and frontend Testing Fixed multiple runtime errors: Indentation issues Missing dataset Database schema mismatches Deployment errors Deployment Hosted on Streamlit Cloud Connected with GitHub for CI/CD ⚠️ Challenges we ran into Deployment issues (missing dependencies like seaborn, sklearn) Database errors (SQLite schema mismatch, locking issues)

Built With

• ci/cd
• github
• mlsckit
• python
• react
• sckit-learn
• seaborn
• streamlit