Introduction to Diffusion Models

MICCAI 2024 MEC Submission6 Authors

15 Aug 2024 (modified: 18 Aug 2024)MICCAI 2024 MEC SubmissionEveryoneRevisionsBibTeXCC BY 4.0
Keywords: Generative Deep Learning, Denoising Diffusion Probabilistic Models, X-ray Imaging
TL;DR: A gentle introduction to denoising diffusion models in medical image analysis, using X-ray imaging as a use-case
Abstract: Our tutorial on denoising diffusion models for medical imaging offers a comprehensive and interactive learning experience. It begins with introductory slides that break down the mathematics behind diffusion models and their training process, making complex concepts accessible. To engage learners, we include an interactive game that demonstrates the potential of diffusion models in medical imaging. Additionally, presentation notes are provided to support lecturers in their teaching and assist students in following along. The tutorial is complemented by a detailed Jupyter notebook, which guides participants through training a diffusion model using the RSNA bone age dataset of X-ray imaging. This tutorial is designed to enhance understanding and application of diffusion models in the medical imaging domain. The link to the Jupyter Notebook is provided in the lecture slides.
Submission Number: 6
Loading