Health Shield AI

Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for Health Shield AI

🧠 Inspiration

Many people around us don’t get regular health checkups due to busy schedules, lack of awareness, or healthcare access. I wanted to build something simple, accessible, and fast that could help users know their health risks—especially related to heart disease and diabetes—just by entering a few basic details.

🛠️ What I Built

Health Shield AI is a lightweight web app built using Python Flask that uses trained Logistic Regression models to predict the risk of:

• 🩺 Diabetes
• ❤️ Heart Disease

The user inputs basic health metrics like age, BMI, glucose level, and blood pressure, and instantly receives a prediction along with a health tip.

🧪 How I Built It

• 📌 Frontend: HTML, basic CSS
• 🧠 Backend: Python Flask
• 🔍 ML Models: Trained with scikit-learn using sample health datasets
• 💾 Model Saving: Used joblib to save and load .pkl models
• 📦 Deployment-ready structure for local use or further cloud hosting

🌱 What I Learned

• How to train and evaluate logistic regression models
• Flask routing and handling HTML forms
• Saving and loading models with joblib
• How to connect AI models with a web interface
• Real-world application of preventive healthcare AI

⚔️ Challenges Faced

• Finding good sample health datasets for quick prototyping
• Fine-tuning the logistic regression models to avoid underfitting
• Integrating the model prediction into a smooth web experience
• Keeping the UI minimal but informative

✅ Final Thoughts

This was a rewarding experience that made me confident in building AI tools for health and social impact. I plan to expand this further with more health predictions and better UI in the future.

Built With

• flask
• html5
• javascript
• joblib
• scikit-learn
• tailwind