Emif8t/Prostate-Cancer-ML-model

This project aims to develop machine learning models for prostate cancer predictions in Africa. # Prostate Cancer Machine Learning Model A modular Python package for developing and evaluating supervised machine learning models for prostate cancer prediction using clinical and molecular biomarkers. The package was developed as part of a PhD research project investigating molecular markers for prostate cancer in African men. It implements a reproducible machine learning workflow including data preprocessing, nested cross-validation, hyperparameter tuning, model evaluation, explainable artificial intelligence (XAI), visualization, and result export. --- ## Features * Modular and object-oriented architecture * Clinical, molecular, and combined feature modelling * Nested stratified cross-validation * Hyperparameter optimization using GridSearchCV * Bootstrap confidence intervals for ROC AUC * Comprehensive performance metrics * ROC and calibration curve visualization * Explainability using SHAP * Permutation and tree-based feature importance * Export of predictions and evaluation metrics * Easily extensible for additional datasets and models --- ## Machine Learning Models The package currently supports: * Logistic Regression * Decision Tree * Random Forest * Extra Trees * Support Vector Machine (SVM) * Extreme Gradient Boosting (XGBoost) --- ## Project Structure ```text Prostate-Cancer-ML-model/ │ ├── data/ │ └── ROCdata2.xlsx │ ├── src/ │ ├── __init__.py │ ├── bootstrap.py │ ├── config.py │ ├── dataset.py │ ├── explainer.py │ ├── exporter.py │ ├── metrics.py │ ├── models.py │ ├── nested_cv.py │ ├── pipeline.py │ ├── preprocess.py │ ├── utils.py │ └── visualization.py │ ├── main.py ├── requirements.txt ├── README.md └── LICENSE ``` --- ## Installation Clone the repository: ```bash git clone github.com ``` Move into the project directory: ```bash cd Prostate-Cancer-ML-model ``` Install the required packages: ```bash pip install -r requirements.txt ``` --- ## Required …