Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

CriDiff: Criss-cross Injection Diffusion Framework via Generative Pre-train for Prostate Segmentation

Tingwei Liu, Miao Zhang, Leiye Liu, Jialong Zhong, Shuyao Wang, Yongri Piao, Huchuan Lu

TL;DR

CriDiff, a two-stage feature injecting framework with a Crisscross Injection Strategy (CIS) and a Generative Pre-train (GP) approach for prostate segmentation that achieves state-of-the-art performance on four evaluation metrics is proposed.

Abstract

Recently, the Diffusion Probabilistic Model (DPM)-based methods have achieved substantial success in the field of medical image segmentation. However, most of these methods fail to enable the diffusion model to learn edge features and non-edge features effectively and to inject them efficiently into the diffusion backbone. Additionally, the domain gap between the images features and the diffusion model features poses a great challenge to prostate segmentation. In this paper, we proposed CriDiff, a two-stage feature injecting framework with a Crisscross Injection Strategy (CIS) and a Generative Pre-train (GP) approach for prostate segmentation. The CIS maximizes the use of multi-level features by efficiently harnessing the complementarity of high and low-level features. To effectively learn multi-level of edge features and non-edge features, we proposed two parallel conditioners in the CIS: the Boundary Enhance Conditioner (BEC) and the Core Enhance Conditioner (CEC), which discriminatively model the image edge regions and non-edge regions, respectively. Moreover, the GP approach eases the inconsistency between the images features and the diffusion model without adding additional parameters. Extensive experiments on four benchmark datasets demonstrate the effectiveness of the proposed method and achieve state-of-the-art performance on four evaluation metrics.

CriDiff: Criss-cross Injection Diffusion Framework via Generative Pre-train for Prostate Segmentation

TL;DR

CriDiff, a two-stage feature injecting framework with a Crisscross Injection Strategy (CIS) and a Generative Pre-train (GP) approach for prostate segmentation that achieves state-of-the-art performance on four evaluation metrics is proposed.

Abstract

Recently, the Diffusion Probabilistic Model (DPM)-based methods have achieved substantial success in the field of medical image segmentation. However, most of these methods fail to enable the diffusion model to learn edge features and non-edge features effectively and to inject them efficiently into the diffusion backbone. Additionally, the domain gap between the images features and the diffusion model features poses a great challenge to prostate segmentation. In this paper, we proposed CriDiff, a two-stage feature injecting framework with a Crisscross Injection Strategy (CIS) and a Generative Pre-train (GP) approach for prostate segmentation. The CIS maximizes the use of multi-level features by efficiently harnessing the complementarity of high and low-level features. To effectively learn multi-level of edge features and non-edge features, we proposed two parallel conditioners in the CIS: the Boundary Enhance Conditioner (BEC) and the Core Enhance Conditioner (CEC), which discriminatively model the image edge regions and non-edge regions, respectively. Moreover, the GP approach eases the inconsistency between the images features and the diffusion model without adding additional parameters. Extensive experiments on four benchmark datasets demonstrate the effectiveness of the proposed method and achieve state-of-the-art performance on four evaluation metrics.
Paper Structure (12 sections, 6 equations, 4 figures, 4 tables)

This paper contains 12 sections, 6 equations, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Generative Pre-train Approach
  4. Boundary Enhance Conditioner and Core Enhance Conditioner
  5. Crisscross Injection Strategy
  6. Experiments
  7. Experiment Protocol
  8. Comparison with State-of-the-arts
  9. Ablation Study
  10. Conclusion
  11. Acknowledgments.
  12. Disclosure of Interests.

Figures (4)

  • Figure 1: Overall of our method. Up: The first stage is our proposed generative pre-train that is described in Sec. 2.1. Bottom: After pre-training, we performed the criss-cross injection strategy to segment prostate in Sec. 2.2 and Sec. 2.3.
  • Figure 2: Detailed structures of our proposed BEC and CEC, which focus on learning the boundary and core information of the prostate under the guidance of decoupled soft labels.
  • Figure 3: Visual comparisons of the proposed model and existing SOTA methods.
  • Figure 4: Injection strategy comparison.