AI-powered medical condition predictor using BERT and TF-IDF models with drug recommendations - Master's Capstone Project

This project applies machine learning techniques to predict cardiovascular disease using clinical data. It focuses on data preprocessing, feature engineering, and supervised learning models to support early diagnosis and clinical decision-making in healthcare.

🫀 Machine Learning web application predicting heart disease risk using 13 medical parameters. Built with HTML/CSS/JS frontend and Python ML backend (85.2% accuracy). Features interactive step-by-step form, real-time predictions, comprehensive data insights, and responsive design. Educational tool with UCI Cleveland dataset.

Diabetes is a growing global health challenge, often undiagnosed until advanced stages. This project uses ML models for early prediction, focusing on data preprocessing, feature engineering, model evaluation, and explainable AI to improve early detection and patient outcomes.

GlucoPredict-AI es una aplicación web que integra inteligencia artificial con datos clínicos y conductuales para estimar el riesgo de desarrollar diabetes tipo 2. Ofrece una interfaz intuitiva para registrar información personal y clínica, generando análisis claros, explicaciones comprensibles y recomendaciones personalizadas.

This project presents a machine learning based system for early detection of diabetes. The application predicts the likelihood of diabetes based on patient health parameters such as glucose level, BMI, blood pressure, insulin level, and other medical indicators.