Overview
Early prediction system for stroke risk based on clinical health profiles, trained on 5,110 medical records across 10 critical clinical features (hypertension, heart disease, glucose levels, BMI, etc.).
Metrics
| Model | Precision | Recall | F1-Score | Accuracy |
|---|---|---|---|---|
| SVM (RBF) | 0.95 | 0.91 | 0.93 | 92.71% |
| Random Forest | 0.93 | 0.90 | 0.91 | 91.06% |
| KNN | 0.97 | 0.79 | 0.87 | 87.95% |
Tech Stack: Python, Scikit-learn, Streamlit, Pandas, NumPy
Last updated on September 16, 2026 at 12:23 PM UTC+7. See Changelog