This project leverages AI and ML models to provide various health-related functionalities, such as health risk prediction, medical text extraction, nutrition estimation, emotion detection, and more.
| Feature | ML Model Used | Description |
|---|---|---|
| Chatbot (AI Assistant) | Transformer Models (GPT, BERT, Rasa NLU) | Provides health advice and answers queries using NLP. |
| OCR (Extracting Text from Images) | Tesseract OCR, EasyOCR, CRNN | Converts medical documents, prescriptions, or handwritten notes into digital text. |
| Heart Risk Prediction | Logistic Regression, SVM, Random Forest, ANNs | Predicts heart disease risk based on user's health data. |
| Step Count & Activity Tracking | LSTM, CNN | Uses wearable sensor data to count steps and detect activity type. |
| Calorie & Nutrition Estimation | YOLO, Faster R-CNN | Recognizes food items from images and estimates calories. |
| Medical Report Analysis | BERT, BioBERT, LLM-based NLP Models | Extracts key insights from medical reports using NLP. |
| Emotion & Stress Detection | CNN, LSTM | Analyzes facial expressions or voice tone to detect stress levels. |
| Sleep Pattern Monitoring | RNN, LSTM | Tracks and predicts sleep quality using time-series analysis. |
| Diabetes & Blood Sugar Prediction | Random Forest, XGBoost, Neural Networks | Predicts diabetes risk from past health records. |
| Medicine Reminder (Voice & Text Alerts) | Speech-to-Text (DeepSpeech), TTS models | Converts text reminders into voice alerts for medicine schedules. |
- Frontend: React.js / Flutter
- Backend: Python (Flask / FastAPI / Django)
- Machine Learning: TensorFlow, PyTorch, Scikit-learn, OpenCV
- Database: PostgreSQL / Firebase / MongoDB
- Cloud & Deployment: AWS / Google Cloud / Docker