Primary Breast Cancer Clinical Dataset from Epic Hospital (Breast_Cancer_METABRIC_Epic_Hospital)
Description
This dataset contains clinical features of breast tumors, designed for use in machine learning and AI-based classification tasks. Each row represents a patient sample with numeric attributes describing tumor characteristics. The dataset includes labels for diagnosis (benign vs malignant), making it suitable for supervised learning tasks such as classification, feature selection, and explainable AI studies. This dataset contains 15,054 breast cancer patient records collected from Epic Hospital. It provides a comprehensive set of clinical, demographic, and pathological features relevant to breast cancer prognosis and treatment analysis. The dataset includes 34 columns, such as patient demographics (age, sex), clinical details (tumor size, tumor stage, grade, histologic subtype), treatment history (surgery, chemotherapy, hormone therapy, radiotherapy), and survival outcomes (overall survival months, overall survival status, relapse status). The dataset is intended to support machine learning, statistical modeling, and medical research in breast cancer survival prediction, treatment response analysis, and outcome assessment. The dataset may be applied to: => Predictive modeling for early breast cancer detection => Feature importance and interpretability studies => Benchmarking traditional ML models vs. LLM-based tabular learning approaches. Researchers and practitioners may use this dataset for academic, educational, or applied AI purposes.
Files
Steps to reproduce
(1) Download the dataset file Breast_cancer_dataset.csv from this repository. (2) Open the file using any spreadsheet software (e.g., Microsoft Excel, Google Sheets) or load it into Python/R for analysis. (3) Each row corresponds to a patient case, and each column represents a clinical feature (such as tumor size, texture, etc.). (4) The diagnosis column indicates whether the tumor is benign (0) or malignant (1). (5) Researchers can directly apply machine learning models (e.g., logistic regression, decision trees, random forest, deep learning) for classification tasks. (6) Preprocessing (e.g., normalization or scaling) may be required before training models.
Institutions
- University of Dhaka Faculty of Science