Data Augmentation for Medical Imaging: Counterfactual Simulation of Acquisition Parameters via Conditional Diffusion Model

Pedro A. Morão, Yasna Forghani, Nuno Loução, Pedro Gouveia, Mario A. T. Figueiredo, Joao Santinha

Keywords: Denoising Diffusion Generative Models, Data Augmentation, MRI, Medical Imaging, Generalizability

TL;DR: This paper introduces a novel method for data augmentation, simulating variations in MRI acquisition parameters without changing patient anatomy. This improves model generalizability, particularly in OOD as shown on the segmentation of breast MRIs.

Abstract: Deep learning (DL) models in medical imaging face challenges in generalizability and robustness due to variations in image acquisition parameters (IAP). In this work, we introduce a novel method using conditional denoising diffusion generative models (cDDGMs) to generate counterfactual medical images that simulate different IAP without altering patient anatomy. We demonstrate that using these counterfactual images for magnetic resonance (MR) data augmentation can improve segmentation accuracy in out-of-distribution settings, enhancing the overall generalizability and robustness of DL models across diverse imaging conditions. Our approach shows promise in addressing domain and covariate shifts in medical imaging. The code is publicly available at https://github.com/pedromorao/Counterfactual-MRI-Data-Augmentation

Primary Subject Area: Generative Models

Secondary Subject Area: Transfer Learning and Domain Adaptation

Paper Type: Both

Registration Requirement: Yes

Reproducibility: https://github.com/pedromorao/Counterfactual-MRI-Data-Augmentation

Visa & Travel: Yes

Midl Latex Submission Checklist: Ensure no LaTeX errors during compilation., Created a single midl25_NNN.zip file with midl25_NNN.tex, midl25_NNN.bib, all necessary figures and files., Includes \documentclass{midl}, \jmlryear{2025}, \jmlrworkshop, \jmlrvolume, \editors, and correct \bibliography command., Did not override options of the hyperref package, Did not use the times package., All authors and co-authors are correctly listed with proper spelling and avoid Unicode characters., Author and institution details are de-anonymized where needed. All author names, affiliations, and paper title are correctly spelled and capitalized in the biography section., References must use the .bib file. Did not override the bibliographystyle defined in midl.cls. Did not use \begin{thebibliography} directly to insert references., Tables and figures do not overflow margins; avoid using \scalebox; used \resizebox when needed., Included all necessary figures and removed *unused* files in the zip archive., Removed special formatting, visual annotations, and highlights used during rebuttal., All special characters in the paper and .bib file use LaTeX commands (e.g., \'e for é)., Appendices and supplementary material are included in the same PDF after references., Main paper does not exceed 9 pages; acknowledgements, references, and appendix start on page 10 or later.

Submission Number: 261