What is RISE?¶
RISE-MICCAI is a dedicated initiative to increase the representation and participation of researchers from under-represented regions in the Medical Image Computing and Computer-Aided Interventions (MICCAI) community — including Latin America, South/Southeast Asia, Africa, and the Middle East.
Our Goals¶
Promote participation from under-represented regions in MICCAI conferences and research initiatives.
Offer mentorship, funding opportunities, and community-driven support for researchers in LMICs.
Cultivate emerging talents in LMICs, helping them gain visibility and recognition in the field.
Foster collaborations across regions, institutions, and continents to address global disparities in medical imaging.
Provide freely accessible, high-quality tutorials in AI for medical imaging to democratize education.
Cover the full spectrum from classical CNNs to the latest foundation models and diffusion architectures.
Tutorials¶
All tutorials are designed to be run interactively. Each notebook includes Google Colab and Binder launch buttons so you can run them in the cloud with zero setup.
An end-to-end deep learning classification pipeline for medical images. Covers CNN architectures, data augmentation, training, and evaluation.
Detect diabetic retinopathy from retinal fundus images using CNNs and pretrained models. Includes class activation maps (CAMs) for interpretability.
Train an attention-based MIL model to detect breast cancer metastases from Whole Slide Images (WSIs) using the CAMELYON16 dataset and the torchmil library.
Covered Topics¶
Across current and upcoming tutorials, the hub covers:
| Category | Topics |
|---|---|
| Classification | CNNs, ResNet, EfficientNet, Transfer Learning |
| Segmentation | U-Net, nnU-Net, pixel-wise labelling |
| Interpretability | Grad-CAM, CAM, SHAP, LIME |
| MIL / Pathology | Attention MIL, CAMELYON, WSI pipelines |
| Self-Supervised | Contrastive Learning, MAE |
| 3D & Video | Volumetric CNNs, 3D Transformers |
| LLMs & Generative | Report generation, Diffusion models, GANs |
| Fairness & Bias | Dataset bias, fairness metrics, mitigation |