Reconstruct or Generate: Exploring the Spectrum of Generative Modeling for Cardiac MRI
By: Niklas Bubeck , Yundi Zhang , Suprosanna Shit and more
Potential Business Impact:
Makes medical pictures clearer and creates fake ones.
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.
Similar Papers
Perceptual Evaluation of GANs and Diffusion Models for Generating X-rays
CV and Pattern Recognition
Makes fake X-rays help doctors find sickness.
Phenotype-Guided Generative Model for High-Fidelity Cardiac MRI Synthesis: Advancing Pretraining and Clinical Applications
CV and Pattern Recognition
Creates fake heart scans to train AI.
Predicting before Reconstruction: A generative prior framework for MRI acceleration
CV and Pattern Recognition
Makes MRI scans faster by predicting images.