# CardioDiT: Latent Diffusion Transformers for 4D Cardiac MRI Synthesis

Source: [CardioDiT: Latent Diffusion Transformers for 4D Cardiac MRI Synthesis](<https://www.sota2.com/research/paper/69c60f000636e5bf4afc84ed>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-28T05:00:04.914Z

Authors: Marvin Seyfarth, Sarah Kaye M\\"uller, Arman Ghanaat, Isabelle Ayx, Fabian Fastenrath, Philipp Wild, Alexander Hertel, Theano Papavassiliu, Salman Ul Hassan Dar, Sandy Engelhardt

Published: 2026-03-26T08:59:07.000Z

## Abstract

Latent diffusion models (LDMs) have recently achieved strong performance in 3D medical image synthesis. However, modalities like cine cardiac MRI (CMR), representing a temporally synchronized 3D volume across the cardiac cycle, add an additional dimension that most generative approaches do not model directly. Instead, they factorize space and time or enforce temporal consistency through auxiliary mechanisms such as anatomical masks. Such strategies introduce structural biases that may limit global context integration and lead to subtle spatiotemporal discontinuities or physiologically inconsistent cardiac dynamics. We investigate whether a unified 4D generative model can learn continuous cardiac dynamics without architectural factorization. We propose CardioDiT, a fully 4D latent diffusion framework for short-axis cine CMR synthesis based on diffusion transformers. A spatiotemporal VQ-VAE encodes 2D+t slices into compact latents, which a diffusion transformer then models jointly as complete 3D+t volumes, coupling space and time throughout the generative process. We evaluate CardioDiT on public CMR datasets and a larger private cohort, comparing it to baselines with progressively stronger spatiotemporal coupling. Results show improved inter-slice consistency, temporally coherent motion, and realistic cardiac function distributions, suggesting that explicit 4D modeling with a diffusion transformer provides a principled foundation for spatiotemporal cardiac image synthesis. Code and models trained on public data are available at https://github.com/Cardio-AI/cardiodit.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [CMR Generation on ACDC and M&Ms-2](<https://www.sota2.com/research/sota/cmr-generation-on-acdc-and-m-and-ms-2.md>): CMR Generation; ACDC and M&Ms-2; FID: 71.9.

- [CMR Generation on Private Dataset](<https://www.sota2.com/research/sota/cmr-generation-on-private-dataset.md>): CMR Generation; Private Dataset; FID: 21.2.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.25194>)

[PDF](<https://arxiv.org/pdf/2603.25194.pdf>)
