Diff-UNet: A Diffusion Embedded Network for Volumetric Segmentation

TL;DR

Diff-UNet introduces a diffusion-embedded U-Net architecture with a Step-Uncertainty Fusion module, significantly improving volumetric segmentation accuracy in medical imaging across multiple datasets.

Contribution

This work integrates diffusion models into a U-Net framework for medical segmentation and proposes a novel fusion module to enhance robustness, which is a new approach in this domain.

Findings

Outperforms state-of-the-art segmentation methods

Effective across diverse medical imaging datasets

Demonstrates universality and robustness of the approach

Abstract

In recent years, Denoising Diffusion Models have demonstrated remarkable success in generating semantically valuable pixel-wise representations for image generative modeling. In this study, we propose a novel end-to-end framework, called Diff-UNet, for medical volumetric segmentation. Our approach integrates the diffusion model into a standard U-shaped architecture to extract semantic information from the input volume effectively, resulting in excellent pixel-level representations for medical volumetric segmentation. To enhance the robustness of the diffusion model's prediction results, we also introduce a Step-Uncertainty based Fusion (SUF) module during inference to combine the outputs of the diffusion models at each step. We evaluate our method on three datasets, including multimodal brain tumors in MRI, liver tumors, and multi-organ CT volumes, and demonstrate that Diff-UNet outperforms other state-of-the-art methods significantly. Our experimental results also indicate the universality and effectiveness of the proposed model. The proposed framework has the potential to facilitate the accurate diagnosis and treatment of medical conditions by enabling more precise segmentation of anatomical structures. The codes of Diff-UNet are available at https://github.com/ge-xing/Diff-UNet