# UniSegDiff: Boosting Unified Lesion Segmentation via a Staged Diffusion Model

arXiv: https://arxiv.org/abs/2507.18362 · HTML: https://mlsanity.com/paper/2507.18362/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models.
- Builds on 19 of its 32 references that are in the map.
- First posted on arXiv 24 Jul 2025.

## Placed in

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

## Builds on (most cited first, 19 of 19)

- [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)
- [TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation](https://mlsanity.com/paper/2102.04306/) (2021)
- [Kvasir-SEG: A Segmented Polyp Dataset](https://mlsanity.com/paper/1911.07069/) (2019)
- [PraNet: Parallel Reverse Attention Network for Polyp Segmentation](https://mlsanity.com/paper/2006.11392/) (2020)
- [SegDiff: Image Segmentation with Diffusion Probabilistic Models](https://mlsanity.com/paper/2112.00390/) (2021)
- [A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Images](https://mlsanity.com/paper/1612.00799/) (2016)
- [MedNeXt: Transformer-driven Scaling of ConvNets for Medical Image Segmentation](https://mlsanity.com/paper/2303.09975/) (2023)
- [MedSegDiff-V2: Diffusion based Medical Image Segmentation with Transformer](https://mlsanity.com/paper/2301.11798/) (2023)
- [One-Step Image Translation with Text-to-Image Models](https://mlsanity.com/paper/2403.12036/) (2024)
- [Ambiguous Medical Image Segmentation using Diffusion Models](https://mlsanity.com/paper/2304.04745/) (2023)
- [EMCAD: Efficient Multi-scale Convolutional Attention Decoding for Medical Image Segmentation](https://mlsanity.com/paper/2405.06880/) (2024)
- [EGE-UNet: an Efficient Group Enhanced UNet for skin lesion segmentation](https://mlsanity.com/paper/2307.08473/) (2023)
- [Make-A-Volume: Leveraging Latent Diffusion Models for Cross-Modality 3D Brain MRI Synthesis](https://mlsanity.com/paper/2307.10094/) (2023)
- [Multi-Scale Coarse-to-Fine Segmentation for Screening Pancreatic Ductal Adenocarcinoma](https://mlsanity.com/paper/1807.02941/) (2018)
- [Stable Diffusion Segmentation for Biomedical Images with Single-step Reverse Process](https://mlsanity.com/paper/2406.18361/) (2024)
- [EBHI-Seg: A Novel Enteroscope Biopsy Histopathological Haematoxylin and Eosin Image Dataset for Image Segmentation Tasks](https://mlsanity.com/paper/2212.00532/) (2022)
- [Conditional diffusion model with spatial attention and latent embedding for medical image segmentation](https://mlsanity.com/paper/2502.06997/) (2025)

## Built on by (most cited first, 2 of 2)

- [Externally Validated Breast Ultrasound Segmentation via Multi-task Learning with BI-RADS-Consistent Morphological Priors](https://mlsanity.com/paper/2511.15968/) (2025)
- [Rethinking Diffusion Segmentation: When Does It Rely on Its Noisy State, and Does Diffusion Matter?](https://mlsanity.com/paper/2609.23967/) (2026)

## Abstract (arXiv)

The Diffusion Probabilistic Model (DPM) has demonstrated remarkable performance across a variety of generative tasks. The inherent randomness in diffusion models helps address issues such as blurring at the edges of medical images and labels, positioning Diffusion Probabilistic Models (DPMs) as a promising approach for lesion segmentation. However, we find that the current training and inference strategies of diffusion models result in an uneven distribution of attention across different timesteps, leading to longer training times and suboptimal solutions. To this end, we propose UniSegDiff, a novel diffusion model framework designed to address lesion segmentation in a unified manner across multiple modalities and organs. This framework introduces a staged training and inference approach, dynamically adjusting the prediction targets at different stages, forcing the model to maintain high attention across all timesteps, and achieves unified lesion segmentation through pre-training the feature extraction network for segmentation. We evaluate performance on six different organs across various imaging modalities. Comprehensive experimental results demonstrate that UniSegDiff significantly outperforms previous state-of-the-art (SOTA) approaches. The code is available at https://github.com/HUYILONG-Z/UniSegDiff.