The Problem
Machine learning models often remain isolated in Jupyter notebooks, making them inaccessible to end-users who need practical, UI-driven predictive insights.
The Solution
Developed an end-to-end machine learning project that trains a LinearRegression model on patient features (cholesterol, age, blood pressure), exports the model via joblib, and serves it through a clean Flask web front-end.
Engineering Challenges
Pipeline Integration: Bridging the gap between data science (scikit-learn training pipelines) and web engineering (Flask REST endpoints).
User Experience: Translating complex probabilistic ML outputs into tailored, readable recommendations for users.
Technology Stack
Data Science & Backend
- Python
- scikit-learn
- Flask
- joblib
Frontend
- HTML
- CSS
- JavaScript
Project Gallery
Image missing: https://raw.githubusercontent.com/kavindu-kodikara/HeartGuard/master/PatientDetails.png
Results & Impact
- Demonstrated the ability to not just train AI models, but successfully deploy them into production-ready web interfaces.
- Showcased proficiency in Python-based full-stack architectures.