Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
October 20, 2025Open Access

Cycle Diffusion Model for Counterfactual Image Generation

View Full Paper
Ask AI
Bookmark
Share

Authors

FHFu-Chu HuangAWAlan WangBLBinxu Li

Discussion

Loading...

Member takes

Overview

Cycle training improves conditioning adherence and image realism in 3D brain MRI datasets, suggesting enhanced counterfactual generation.

Key Points

  • Enhanced image quality is achieved through the cycle diffusion model, improving conditioning accuracy for medical imaging.
  • The cycle training framework outperforms previous methods, resulting in better synthetic images measured by FID and SSIM.
  • Experiments on 3D brain MRI datasets reveal that the proposed model supports reliable direct and counterfactual image generation.
  • This approach highlights the potential for refining medical image synthesis, specifically in data augmentation and disease modeling.

Cite This Study

Huang et al. (2025) studied this question.

synapsesocial.com/papers/68f5fcce8d54a28a75cf1bbahttps://doi.org/10.48550/arxiv.2509.24267
View Full Paper
Ask AI
Bookmark
Share