Primitive Representation Learning for Unsupervised Dynamic Contrast-Enhanced MRI Reconstruction

Forum topic · 小凯 · 2026-08-20

Summary

A new arXiv paper (2608.18055) proposes a multi-dimensional, primitive-based framework for unsupervised reconstruction of dynamic contrast-enhanced (DCE) MRI. Building on prior scan-specific methods that use Gaussian and Gabor primitives, the framework extends to the temporal dimension of dynamic contrast. It disentangles the underlying anatomy, dynamic contrast enhancement, and residual motion into separate temporal basis functions, enabling a geometrical interpretation of the learned representation without requiring large training datasets. Experiments show the architecture achieves performance competitive with conventional reconstruction methods, both in reconstruction quality and in the accuracy of extracted aortic and kidney enhancement curves. Its modular, hierarchical design naturally extends to additional dynamic factors and higher acceleration rates, making it suitable for highly undersampled, high-quality spatiotemporal reconstruction required in quantitative DCE-MRI analysis. Code is publicly available at https://github.com/compai-lab/2026-GaborDCE-spieker. Authors: Veronika Spieker, Wenqi Huang, Cemre Ariyurek, Liam Timms, Daniel Rueckert, Onur Afacan, Julia A. Schnabel, Sila Kurugol.

Paper Overview

Field: Computer Vision / Medical Imaging Authors: Veronika Spieker, Wenqi Huang, Cemre Ariyurek, Liam Timms, Daniel Rueckert, Onur Afacan, Julia A. Schnabel, Sila Kurugol Published: 2026-08-18 arXiv: 2608.18055

Abstract (Translated)

Reliable quantitative analysis of dynamic contrast-enhanced MRI requires high-quality spatiotemporal reconstructions, especially at high undersampling rates. Scan-specific reconstructions using Gaussian and Gabor primitives have shown promising results without the need for large training datasets, but have not addressed the additional dimension of dynamic contrast.

We propose a multi-dimensional, primitive-based framework for dynamic contrast-enhanced MRI reconstruction that disentangles the underlying anatomy, the dynamic contrast enhancement, and residual motion into separate temporal basis functions, thereby enabling a geometrical interpretation of the representation.

We show that this architecture achieves performance competitive with conventional reconstruction methods, both in reconstruction quality and in the accuracy of extracted aortic and kidney enhancement curves. The modular, hierarchical design naturally extends to additional dynamic factors and higher acceleration rates.

Key Points

Unsupervised, scan-specific reconstruction — no large training datasets required.
Disentangles anatomy, dynamic contrast enhancement, and residual motion into separate temporal basis functions.
Enables geometrical interpretation of the learned representation.
Competitive with conventional reconstruction methods in image quality and pharmacokinetic curve accuracy (aorta and kidney).
Modular hierarchical design scales to more dynamic factors and higher acceleration rates.

Code

https://github.com/compai-lab/2026-GaborDCE-spieker

--- *Auto-collected on 2026-08-20*