# MSRepaint: Multiple Sclerosis Repaint with Conditional Denoising Diffusion Implicit Model for Bidirectional Lesion Filling and Synthesis

arXiv: https://arxiv.org/abs/2510.02063 · HTML: https://mlsanity.com/paper/2510.02063/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models.
- Builds on 21 of its 57 references that are in the map.
- First posted on arXiv 2 Oct 2025.

## Placed in

- Diffusion models / [Medical imaging and segmentation](https://mlsanity.com/diffusion/medical-imaging-and-segmentation/)

## Builds on (most cited first, 21 of 21)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Improved Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2102.09672/) (2021)
- [Which Training Methods for GANs do actually Converge?](https://mlsanity.com/paper/1801.04406/) (2018)
- [VoxelMorph: A Learning Framework for Deformable Medical Image Registration](https://mlsanity.com/paper/1809.05231/) (2018)
- [Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation](https://mlsanity.com/paper/2012.07177/) (2020)
- [FastSurfer -- A fast and accurate deep learning based neuroimaging pipeline](https://mlsanity.com/paper/1910.03866/) (2019)
- [3D Whole Brain Segmentation using Spatially Localized Atlas Network Tiles](https://mlsanity.com/paper/1903.12152/) (2019)
- [CarveMix: A Simple Data Augmentation Method for Brain Lesion Segmentation](https://mlsanity.com/paper/2108.06883/) (2021)
- [HACA3: A Unified Approach for Multi-site MR Image Harmonization](https://mlsanity.com/paper/2212.06065/) (2022)
- [Multiple Sclerosis Lesion Synthesis in MRI using an encoder-decoder U-NET](https://mlsanity.com/paper/1901.05733/) (2019)
- [Beyond the LUMIR challenge: The pathway to foundational registration models](https://mlsanity.com/paper/2505.24160/) (2025)
- [TumorCP: A Simple but Effective Object-Level Data Augmentation for Tumor Segmentation](https://mlsanity.com/paper/2107.09843/) (2021)
- [Towards an accurate and generalizable multiple sclerosis lesion segmentation model using self-ensembled lesion fusion](https://mlsanity.com/paper/2312.01460/) (2023)
- [Bi-Directional MS Lesion Filling and Synthesis Using Denoising Diffusion Implicit Model-based Lesion Repainting](https://mlsanity.com/paper/2410.05027/) (2024)
- [A Self-Adaptive Network For Multiple Sclerosis Lesion Segmentation From Multi-Contrast MRI With Various Imaging Protocols](https://mlsanity.com/paper/1811.07491/) (2018)
- [Subject-Specific Lesion Generation and Pseudo-Healthy Synthesis for Multiple Sclerosis Brain Images](https://mlsanity.com/paper/2208.02135/) (2022)
- [Unique MS Lesion Identification from MRI](https://mlsanity.com/paper/2410.09639/) (2024)
- [Encoder-Only Image Registration](https://mlsanity.com/paper/2509.00451/) (2025)
- [UNISELF: A Unified Network with Instance Normalization and Self-Ensembled Lesion Fusion for Multiple Sclerosis Lesion Segmentation](https://mlsanity.com/paper/2508.03982/) (2025)

## Built on by (most cited first, 1 of 1)

- [MetaVoxel: Joint Diffusion Modeling of Imaging and Clinical Metadata](https://mlsanity.com/paper/2512.10041/) (2025)

## Abstract (arXiv)

In multiple sclerosis, lesions interfere with automated magnetic resonance imaging analyses such as brain parcellation and deformable registration, while lesion segmentation models are hindered by the limited availability of annotated training data. To address both issues, we propose MSRepaint, a unified diffusion-based generative model for bidirectional lesion filling and synthesis that restores anatomical continuity for downstream analyses and augments segmentation through realistic data generation. MSRepaint conditions on spatial lesion masks for voxel-level control, incorporates contrast dropout to handle missing inputs, integrates a repainting mechanism to preserve surrounding anatomy during lesion filling and synthesis, and employs a multi-view DDIM inversion and fusion pipeline for 3D consistency with fast inference. Extensive evaluations demonstrate the effectiveness of MSRepaint across multiple tasks. For lesion filling, we evaluate both the accuracy within the filled regions and the impact on downstream tasks including brain parcellation and deformable registration. MSRepaint outperforms the traditional lesion filling methods FSL and NiftySeg, and achieves accuracy on par with FastSurfer-LIT, a recent diffusion model-based inpainting method, while offering over 20 times faster inference. For lesion synthesis, state-of-the-art MS lesion segmentation models trained on MSRepaint-synthesized data outperform those trained on CarveMix-synthesized data or real ISBI challenge training data across multiple benchmarks, including the MICCAI 2016 and UMCL datasets. Additionally, we demonstrate that MSRepaint's unified bidirectional filling and synthesis capability, with full spatial control over lesion appearance, enables high-fidelity simulation of lesion evolution in longitudinal MS progression.