How edaaydinea/ai-projects-for-healthcare works
Project type and purpose
• This repository is not a single deployed service but a monorepo of end-to-end data-science/ML projects in healthcare.
• Each top-level folder contains a standalone Jupyter notebook demonstrating a data ingestion → preprocessing → modeling → evaluation pipeline for a specific healthcare problem (e.g., breast cancer, COVID-19 detection, diabetes prediction).
File structure
• Top-level directories: one per project (e.g., “Breast Cancer Classification,” “Diabetes Prediction With Pyspark MLLIB”)
• Inside each:
• One or more .ipynb notebooks implementing data loading, feature engineering, model training, visualization
• Datasets in .csv, .xlsx or image form
• Occasionally environment files (environment.yml) for reproducibility
• Root: LICENSE, README.md – no frontend/backend separation or API layer, purely analysis pipelines.
README for additional insights
• Confirms each folder is a completed (✅) or work-in-progress analysis project
• Lists dependencies used across projects (PyTorch, PySpark MLLIB, Plotly, Keras/TensorFlow, scikit-learn)
• No mention of microservices or deployments—focus is on notebook-driven experiments.
Detail
• Versioned environment files as a first-class component