MICCAI 2026 Tutorial: Multimodal Foundational Models for BioMedical Imaging (MultiFound)

This repository is for the MICCAI 2026 Tutorial on Multimodal Foundational Models for BioMedical Imaging (MultiFound).
The MICCAI 2026 edition marks the third edition of the MultiFound tutorial series, which was first launched at MICCAI 2024.
🌐 Tutorial website: https://sites.google.com/view/miccai-2026-tutorial

Organizers

• Ismail Ben Ayed — ÉTS Montréal
• Tanveer Syeda-Mahmood — Stanford University
• Yunsoo Kim — University College London
• Chaoyi Wu — Shanghai Jiao Tong University
• Weidi Xie — Shanghai Jiao Tong University
• Honghan Wu — University of Glasgow

About the Tutorial

MultiFound focuses on recent advances in multimodal foundation models for biomedical imaging, bringing together methods that integrate medical images with complementary modalities such as text, clinical information, and other biomedical data.

The tutorial is intended to provide participants with an overview of the field, key methodological concepts, representative applications, and emerging research directions.

Topics

The tutorial will cover topics related to:

• Multimodal foundation models for biomedical imaging
• Vision-language models in medicine
• Medical image and text representation learning
• Multimodal pretraining and alignment
• Adaptation and fine-tuning of foundation models
• Evaluation of multimodal medical AI systems
• Clinical and research applications

Tutorial Program and Slides

The MICCAI 2026 MultiFound tutorial is organized into seven sessions.

• 1. Introduction to Foundation Models slides — 8:00–8:40 AM
• 2. Contrastive Vision-Language Models for Medical Imaging slides — 8:40–9:20 AM
• 3. Fine-tuning and Adaptation of Foundation Models slides — 9:20–10:00 AM
• 4. VLM for Image Generation slides segmentation_slides — 10:30–10:50 AM
• 5. Multimodal LLMs for Text Generation slides — 10:50–11:10 AM
• 6. Medical Multimodal LLMs slides — 11:10 AM–12:20 PM
• 7. Closing Remarks — 12:20–12:30 PM

Slides

The tutorial materials are provided as three presenter-specific slide decks:

• Tanveer — Introduction, VLM for image generation, multimodal LLMs for text generation, and closing remarks part1 part4 part4_segmentation part5
• Ismail — Contrastive vision-language models and fine-tuning/adaptation of foundation models part2 part3
• Yunsoo — Medical multimodal LLMs part6