Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Balancing Fidelity, Utility, and Privacy in Synthetic Cardiac MRI Generation: A Comparative Study

Madhura Edirisooriya, Dasuni Kawya, Ishan Kumarasinghe, Isuri Devindi, Mary M. Maleckar, Roshan Ragel, Isuru Nawinne, Vajira Thambawita

TL;DR

This study systematically benchmarks three generative architectures: Denoising Diffusion Probabilistic Models (DDPM), Latent Diffusion Models (LDM), and Flow Matching for synthetic CMR generation, showing that diffusion-based models provide the most effective balance between downstream segmentation utility, image fidelity, and privacy preservation under limited-data conditions.

Abstract

Deep learning in cardiac MRI (CMR) is fundamentally constrained by both data scarcity and privacy regulations. This study systematically benchmarks three generative architectures: Denoising Diffusion Probabilistic Models (DDPM), Latent Diffusion Models (LDM), and Flow Matching (FM) for synthetic CMR generation. Utilizing a two-stage pipeline where anatomical masks condition image synthesis, we evaluate generated data across three critical axes: fidelity, utility, and privacy. Our results show that diffusion-based models, particularly DDPM, provide the most effective balance between downstream segmentation utility, image fidelity, and privacy preservation under limited-data conditions, while FM demonstrates promising privacy characteristics with slightly lower task-level performance. These findings quantify the trade-offs between cross-domain generalization and patient confidentiality, establishing a framework for safe and effective synthetic data augmentation in medical imaging.

Balancing Fidelity, Utility, and Privacy in Synthetic Cardiac MRI Generation: A Comparative Study

TL;DR

This study systematically benchmarks three generative architectures: Denoising Diffusion Probabilistic Models (DDPM), Latent Diffusion Models (LDM), and Flow Matching for synthetic CMR generation, showing that diffusion-based models provide the most effective balance between downstream segmentation utility, image fidelity, and privacy preservation under limited-data conditions.

Abstract

Deep learning in cardiac MRI (CMR) is fundamentally constrained by both data scarcity and privacy regulations. This study systematically benchmarks three generative architectures: Denoising Diffusion Probabilistic Models (DDPM), Latent Diffusion Models (LDM), and Flow Matching (FM) for synthetic CMR generation. Utilizing a two-stage pipeline where anatomical masks condition image synthesis, we evaluate generated data across three critical axes: fidelity, utility, and privacy. Our results show that diffusion-based models, particularly DDPM, provide the most effective balance between downstream segmentation utility, image fidelity, and privacy preservation under limited-data conditions, while FM demonstrates promising privacy characteristics with slightly lower task-level performance. These findings quantify the trade-offs between cross-domain generalization and patient confidentiality, establishing a framework for safe and effective synthetic data augmentation in medical imaging.
Paper Structure (31 sections, 9 equations, 4 figures, 6 tables)

This paper contains 31 sections, 9 equations, 4 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methodology
  4. Mask Generation
  5. Image Generation
  6. Denoising Diffusion Probabilistic Model
  7. Latent Diffusion Model
  8. Flow Matching
  9. Evaluation methods
  10. Synthetic Mask Evaluation
  11. Synthetic MRI Fidelity Evaluation
  12. Utility Evaluation via Cardiac MRI Segmentation
  13. Synthetic CMR Privacy Evaluation
  14. Nearest Neighbor-based Privacy Analysis
  15. Membership Inference Attack (MIA)
  16. ...and 16 more sections

Figures (4)

  • Figure 1: Overview of the proposed synthetic cardiac MRI generation and evaluation framework. The pipeline consists of two main stages: (1) segmentation mask generation using a diffusion-based model, followed by (2) mask-conditioned image synthesis using DDPM, LDM, and Flow Matching. The generated images are evaluated across three dimensions: fidelity (PSNR, SSIM, MS-SSIM, LPIPS, FID, KID), downstream utility via cross-dataset cardiac MRI segmentation, and privacy through nearest-neighbor analysis and membership inference attacks.
  • Figure 2: Denoising Process. An image is shown each 200 timesteps
  • Figure 3: CMRI synthesis results. Top row: input segmentation masks. Subsequent rows: synthetic images generated by DDPM, LDM, and FM respectively
  • Figure 4: Comparison of geometric shape metrics between real and synthetic cardiac masks. (a) Area reflects the size of the segmented structure, (b) Eccentricity measures elongation, (c) Roundness evaluates circular compactness, and (d) Solidity assesses boundary smoothness and structural integrity. Overall, the real and synthetic distributions show strong alignment across most structures, with only minor deviations in (c) roundness and (d) solidity, indicating that the generated masks largely preserve anatomical geometry as assessed by (a)-(d).