HyperCoil-Recon: A Hypernetwork-based Adaptive Coil Configuration Task Switching Network for MRI Reconstruction

TL;DR

HyperCoil-Recon introduces a hypernetwork-based approach for MRI reconstruction that adaptively handles varying coil configurations, including unseen ones, outperforming traditional models and enabling more flexible clinical applications.

Contribution

The paper presents a novel hypernetwork framework for MRI reconstruction that encodes multiple coil configurations as tasks, allowing adaptive and generalizable performance across diverse and unseen setups.

Findings

Adapts to unseen configurations with up to 32 coils after training on fewer coils.

Matches performance of configuration-specific models.

Outperforms configuration-invariant models in PSNR and SSIM metrics.

Abstract

Parallel imaging, a fast MRI technique, involves dynamic adjustments based on the configuration i.e. number, positioning, and sensitivity of the coils with respect to the anatomy under study. Conventional deep learning-based image reconstruction models have to be trained or fine-tuned for each configuration, posing a barrier to clinical translation, given the lack of computational resources and machine learning expertise for clinicians to train models at deployment. Joint training on diverse datasets learns a single weight set that might underfit to deviated configurations. We propose, HyperCoil-Recon, a hypernetwork-based coil configuration task-switching network for multi-coil MRI reconstruction that encodes varying configurations of the numbers of coils in a multi-tasking perspective, posing each configuration as a task. The hypernetworks infer and embed task-specific weights into the reconstruction network, 1) effectively utilizing the contextual knowledge of common and varying image features among the various fields-of-view of the coils, and 2) enabling generality to unseen configurations at test time. Experiments reveal that our approach 1) adapts on the fly to various unseen configurations up to 32 coils when trained on lower numbers (i.e. 7 to 11) of randomly varying coils, and to 120 deviated unseen configurations when trained on 18 configurations in a single model, 2) matches the performance of coil configuration-specific models, and 3) outperforms configuration-invariant models with improvement margins of around 1 dB / 0.03 and 0.3 dB / 0.02 in PSNR / SSIM for knee and brain data. Our code is available at https://github.com/sriprabhar/HyperCoil-Recon