Semi-Supervised Learning of Mutually Accelerated MRI Synthesis without Fully-Sampled Ground Truths
This work improves the feasibility and utility of accelerated MRI acquisitions for medical imaging by reducing the need for extensive fully-sampled data, which is a significant practical problem for clinicians and researchers.
This paper addresses the challenge of synthesizing multi-contrast MRI images without relying on fully-sampled ground truths, which are often impractical to acquire. The proposed semi-supervised deep generative model learns to recover high-quality target images directly from accelerated acquisitions of both source and target contrasts, achieving performance equivalent to fully-supervised models.
Learning-based synthetic multi-contrast MRI commonly involves deep models trained using high-quality images of source and target contrasts, regardless of whether source and target domain samples are paired or unpaired. This results in undesirable reliance on fully-sampled acquisitions of all MRI contrasts, which might prove impractical due to limitations on scan costs and time. Here, we propose a novel semi-supervised deep generative model that instead learns to recover high-quality target images directly from accelerated acquisitions of source and target contrasts. To achieve this, the proposed model introduces novel multi-coil tensor losses in image, k-space and adversarial domains. These selective losses are based only on acquired k-space samples, and randomized sampling masks are used across subjects to capture relationships among acquired and non-acquired k-space regions. Comprehensive experiments on multi-contrast neuroimaging datasets demonstrate that our semi-supervised approach yields equivalent performance to gold-standard fully-supervised models, while outperforming a cascaded approach that learns to synthesize based on reconstructions of undersampled data. Therefore, the proposed approach holds great promise to improve the feasibility and utility of accelerated MRI acquisitions mutually undersampled across both contrast sets and k-space.