CTS: A Consistency-Based Medical Image Segmentation Model

TL;DR

This paper introduces CTS, a novel consistency-based model for medical image segmentation that achieves high-quality results with only one sampling, significantly improving speed over traditional diffusion models.

Contribution

The paper adapts consistency models for medical image segmentation, designing new supervision modes and loss functions to enable effective single-sampling segmentation.

Findings

CTS achieves better segmentation accuracy with single sampling.

The model significantly speeds up training and prediction.

Experiments validate the effectiveness of the proposed approach.

Abstract

In medical image segmentation tasks, diffusion models have shown significant potential. However, mainstream diffusion models suffer from drawbacks such as multiple sampling times and slow prediction results. Recently, consistency models, as a standalone generative network, have resolved this issue. Compared to diffusion models, consistency models can reduce the sampling times to once, not only achieving similar generative effects but also significantly speeding up training and prediction. However, they are not suitable for image segmentation tasks, and their application in the medical imaging field has not yet been explored. Therefore, this paper applies the consistency model to medical image segmentation tasks, designing multi-scale feature signal supervision modes and loss function guidance to achieve model convergence. Experiments have verified that the CTS model can obtain better medical image segmentation results with a single sampling during the test phase.