This repository contains all hands-on practical exercises for the AI for Medical Diagnosis and Prediction course. Each week's folder includes starter Jupyter notebooks, sample data pipelines, and instructions needed to complete the formative labs.
/
├── environment.yml # Conda environment definition
├── requirements.txt # pip dependencies
├── data/ # Sample data subsets (placeholders)
│ ├── ReMIND/ # Brain MRI DICOMs
│ ├── PKG - ReMIND_NRRD_Seg_Sep_2023/ # Tumor segmentation in NRRD format
│ ├── mimic_cxr/ # Chest X‑ray DICOMs + reports
│ └── mimic_iv/ # EHR CSV snippets
├── notebooks/ # Jupyter notebooks by week
│ ├── week1/
│ └── week2/
│ ├── lesson1.ipynb # Practical exercises for lesson #1 of week #2
│ └── lesson2.ipynb # Practical exercises for lesson #2 of week #2
├── utils/ # Helper scripts
│ ├── data_loader.py # DICOM & CSV loading utilities
│ └── metrics.py # Clinical metric
├── LICENSE
└── README.md # This filegit clone https://github.com/albarqounilab/AIM.git
cd AIMUsing Conda (recommended)
conda env create -f environment.yml
conda activate aim-courseUsing pip
pip install -r requirements.txtjupyter labOpen the notebook corresponding to the week you're working on, and follow the instructions within each notebook.
- utils/data_loader.py: Load and preprocess DICOM images and CSV tables.
- utils/metrics.py: Calculate clinical metrics such as sensitivity, specificity, AUC, and Dice scores.