Electrical Engineering and Systems Science > Image and Video Processing
[Submitted on 7 Feb 2024 (this version), latest version 19 Jun 2024 (v4)]
Title:Anatomically-Controllable Medical Image Generation with Segmentation-Guided Diffusion Models
View PDF HTML (experimental)Abstract:Diffusion models have enabled remarkably high-quality medical image generation, which can help mitigate the expenses of acquiring and annotating new images by supplementing small or imbalanced datasets, along with other applications. However, these are hampered by the challenge of enforcing global anatomical realism in generated images. To this end, we propose a diffusion model for anatomically-controlled medical image generation. Our model follows a multi-class anatomical segmentation mask at each sampling step and incorporates a \textit{random mask ablation} training algorithm, to enable conditioning on a selected combination of anatomical constraints while allowing flexibility in other anatomical areas. This also improves the network's learning of anatomical realism for the completely unconditional (unconstrained generation) case. Comparative evaluation on breast MRI and abdominal/neck-to-pelvis CT datasets demonstrates superior anatomical realism and input mask faithfulness over state-of-the-art models. We also offer an accessible codebase and release a dataset of generated paired breast MRIs. Our approach facilitates diverse applications, including pre-registered image generation, counterfactual scenarios, and others.
Submission history
From: Nicholas Konz [view email][v1] Wed, 7 Feb 2024 19:35:09 UTC (5,962 KB)
[v2] Fri, 9 Feb 2024 21:40:29 UTC (23,105 KB)
[v3] Mon, 11 Mar 2024 17:50:31 UTC (32,310 KB)
[v4] Wed, 19 Jun 2024 21:10:31 UTC (32,306 KB)
Current browse context:
eess.IV
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.