| --- |
| language: en |
| license: mit |
| library_name: scikit-learn |
| tags: |
| - tabular-classification |
| - healthcare |
| - clinical-prediction |
| - risk-stratification |
| - explainable-ai |
| metrics: |
| - roc_auc |
| - accuracy |
| - recall |
| - f1 |
| model-index: |
| - name: heart-attack-risk-classifier |
| results: |
| - task: |
| type: tabular-classification |
| metrics: |
| - type: roc_auc |
| value: 0.912 |
| - type: recall |
| value: 0.923 |
| - type: f1 |
| value: 0.884 |
| --- |
| |
| # Clinical Heart Attack Risk Prediction Model 🫀 |
|
|
| A calibrated tabular gradient boosting classifier trained to predict acute myocardial infarction risk from non-invasive clinical intake markers. |
|
|
| - **Associated Space**: [Clinical Heart Attack Risk Triage Calculator](https://huggingface.co/spaces/Jayan12321/clinical-heart-risk-calculator) |
| - **Author**: [Jayan Gupta](https://huggingface.co/Jayan12321) |
|
|
| --- |
|
|
| ## 🎯 Clinical Sensitivity & Objective |
| In medical triage modeling, false negatives are far more critical than false positives. This classifier is tuned for high sensitivity, achieving **>92% Recall** on cardiac event detection to minimize life-threatening missed diagnoses. |
|
|
| --- |
|
|
| ## 📊 Features Required for Inference |
|
|
| | Feature | Description | Values / Range | |
| |---|---|---| |
| | `age` | Patient age | Years | |
| | `sex` | Biological sex | `1` = male, `0` = female | |
| | `cp` | Chest Pain type | `0`: Typical Angina, `1`: Atypical, `2`: Non-anginal, `3`: Asymptomatic | |
| | `trestbps` | Resting Blood Pressure | mm Hg on admission | |
| | `chol` | Serum Cholesterol | mg/dl | |
| | `fbs` | Fasting Blood Sugar | `1` if > 120 mg/dl, `0` otherwise | |
| | `restecg` | Resting ECG results | `0`: Normal, `1`: ST-T wave abnormality, `2`: Left ventricular hypertrophy | |
| | `thalach` | Maximum Heart Rate achieved | bpm | |
| | `exang` | Exercise Induced Angina | `1` = yes, `0` = no | |
| | `oldpeak` | ST depression induced by exercise | Relative to rest (float) | |
| | `slope` | Slope of peak exercise ST segment | `0`: Upsloping, `1`: Flat, `2`: Downsloping | |
| | `ca` | Number of major vessels colored by flourosopy | `0-3` | |
| | `thal` | Thalassemia status | `1`: Normal, `2`: Fixed defect, `3`: Reversible defect | |
|
|
| --- |
|
|
| ## 💻 Python Quickstart |
|
|
| ```python |
| import joblib |
| from huggingface_hub import hf_hub_download |
| |
| # Download model from Hugging Face |
| model_path = hf_hub_download( |
| repo_id="Jayan12321/heart-attack-risk-classifier", |
| filename="model.joblib" |
| ) |
| |
| # Load classifier |
| model = joblib.load(model_path) |
| |
| # Sample patient features |
| # [age, sex, cp, trestbps, chol, fbs, restecg, thalach, exang, oldpeak, slope, ca, thal] |
| patient_features = [[57, 1, 2, 130, 236, 0, 0, 174, 0, 0.0, 1, 1, 2]] |
| |
| # Predict calibrated risk probability |
| risk_prob = model.predict_proba(patient_features)[0, 1] |
| print(f"Predicted Cardiac Event Probability: {risk_prob:.2%}") |
| ``` |
|
|
| --- |
|
|
| ## ⚠️ Educational Disclaimer |
| *This model is provided solely for educational and research demonstration purposes. It is not approved as a medical diagnostic device and must not be used for actual patient diagnosis.* |
|
|
