Reconstruct or Generate: Exploring the Spectrum of Generative Modeling for Cardiac MRI

TL;DR

This paper systematically analyzes how modern generative models, including diffusion and autoregressive types, perform across cardiac MRI reconstruction and generation tasks, highlighting their strengths and limitations.

Contribution

It introduces a comprehensive benchmark and analysis of generative models' behavior in medical imaging, emphasizing the spectrum between reconstruction and generation.

Findings

Diffusion models excel in perceptual quality for unconditional generation.

Autoregressive models maintain stable performance across masking ratios.

Diffusion models tend to hallucinate with higher masking ratios.

Abstract

In medical imaging, generative models are increasingly relied upon for two distinct but equally critical tasks: reconstruction, where the goal is to restore medical imaging (usually inverse problems like inpainting or superresolution), and generation, where synthetic data is created to augment datasets or carry out counterfactual analysis. Despite shared architecture and learning frameworks, they prioritize different goals: generation seeks high perceptual quality and diversity, while reconstruction focuses on data fidelity and faithfulness. In this work, we introduce a "generative model zoo" and systematically analyze how modern latent diffusion models and autoregressive models navigate the reconstruction-generation spectrum. We benchmark a suite of generative models across representative cardiac medical imaging tasks, focusing on image inpainting with varying masking ratios and sampling strategies, as well as unconditional image generation. Our findings show that diffusion models offer superior perceptual quality for unconditional generation but tend to hallucinate as masking ratios increase, whereas autoregressive models maintain stable perceptual performance across masking levels, albeit with generally lower fidelity.