Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces a pipeline for enhancing ultra-low-field (ULF) MRI images by synthesizing high-field-like MRIs. The approach uses a Swin UNETR to generate tissue segmentation priors from ULF data, which then condition two independent enhancement networks: a CycleGAN and a transformer-based residual enhancement model (T-REX). Combining the outputs of these models via weighted averaging yields enhanced MRIs comparable to high-field scans.
Segmentation-guided adversarial learning can make ultra-low-field MRI quality nearly indistinguishable from high-field scans.
Ultra-low-field (ULF) MRI offers portable and low-cost imaging but suffers from poor image quality. To address this, we present our submission to the 2025 ULF Enhancement Challenge (ULF-EnC), where the goal is to synthesise high-field-like MRIs from 64 mT scans. Our pipeline enhances ULF MRI through a combination of anatomical conditioning and model ensembling. We first generate tissue segmentation priors using a Swin UNETR trained solely on challenge-provided data. These priors condition two independent enhancement networks - a CycleGAN and a transformer-based residual enhancement model (T-REX) - each trained to synthesise 3 T-like MRIs. Outputs from both models are combined using a weighted average. Our approach produces enhanced MRIs that were comparable to high-field scans both quantitatively and qualitatively.