Self-Supervised Learning for Improved Calibrationless Radial MRI with NLINV-Net

• Published in: ArXiv.org 2024-07
• Main Authors: Blumenthal, Moritz, Fantinato, Chiara, Unterberg-Buchwald, Christina, Haltmeier, Markus, Wang, Xiaoqing, Uecker, Martin
• Format: Article
• Language: English
• ISSN: 2331-8422; 2331-8422

To develop a neural network architecture for improved calibrationless reconstruction of radial data when no ground truth is available for training. NLINV-Net is a model-based neural network architecture that directly estimates images and coil sensitivities from (radial) k-space data via non-linear inversion (NLINV). Combined with a training strategy using self-supervision via data undersampling (SSDU), it can be used for imaging problems where no ground truth reconstructions are available. We validated the method for (1) real-time cardiac imaging and (2) single-shot subspace-based quantitative T1 mapping. Furthermore, region-optimized virtual (ROVir) coils were used to suppress artifacts stemming from outside the FoV and to focus the k-space based SSDU loss on the region of interest. NLINV-Net based reconstructions were compared with conventional NLINV and PI-CS (parallel imaging + compressed sensing) reconstruction and the effect of the region-optimized virtual coils and the type of training loss was evaluated qualitatively. NLINV-Net based reconstructions contain significantly less noise than the NLINV-based counterpart. ROVir coils effectively suppress streakings which are not suppressed by the neural networks while the ROVir-based focussed loss leads to visually sharper time series for the movement of the myocardial wall in cardiac real-time imaging. For quantitative imaging, T1-maps reconstructed using NLINV-Net show similar quality as PI-CS reconstructions, but NLINV-Net does not require slice-specific tuning of the regularization parameter. NLINV-Net is a versatile tool for calibrationless imaging which can be used in challenging imaging scenarios where a ground truth is not available.