Early lung cancer risk prediction using machine-learning models trained on the PLCO dataset. End-to-end pipeline including data cleaning, feature preprocessing, multiple ML baselines, super-stacking ensembles, class-imbalance handling, optimal threshold tuning for high recall, comprehensive evaluation, and SHAP-based model interpretability.
python machine-learning neural-network random-forest healthcare xgboost lightgbm ensemble-learning lung-cancer research-project gradient-boosting imbalanced-learning catboost cancer-prediction medical-ai stacking-ensemble risk-prediction plco
-
Updated
Jan 7, 2026 - Jupyter Notebook