Deep learning for restoring MPI system matrices using simulated training data

OBJECTIVE: Magnetic particle imaging reconstructs tracer distributions using a system matrix obtained through time-consuming, noise-prone calibration measurements. Methods for addressing imperfections in measured system matrices increasingly rely on deep neural networks, yet curated training data remain scarce. This study evaluates whether physics-based simulated system matrices can be used to train deep learning models for different system matrix restoration tasks, i.e., denoising, accelerated calibration, upsampling, and inpainting, that generalize to measured data. APPROACH: A large system matrices dataset was generated using an equilibrium magnetization model extended with uniaxial anisotropy. The dataset spans particle, scanner, and calibration parameters for 2D and 3D trajectories, and includes background noise injected from empty-frame measurements. For each restoration task, deep learning models were compared with classical non-learning baseline methods. Quantitative performance was evaluated on simulated data using peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). For measured data, performance was assessed qualitatively by visual comparison of system matrices and the resulting reconstructions. MAIN RESULTS: The models trained solely on simulated system matrices generalized to measured data across all tasks: for denoising, DnCNN/RDN/SwinIR outperformed DCT-F baseline by > 10 dB PSNR and up to +0.1 SSIM on simulations and led to perceptually better reconstuctions of real data; for 2D upsampling, SMRnet exceeded bicubic by ∼ 20 dB PSNR and ∼ 0.08 SSIM at ×2-×4 which did not transfer qualitatively to real measurements. For 3D accelerated calibration, SMRnet matched tricubic in noiseless cases and was more robust under noise, and for 3D inpainting, biharmonic inpainting was superior when noise-free but degraded with noise, while a PConvUNet maintained quality and yielded less blurry reconstructions. SIGNIFICANCE: The demonstrated transferability of deep learning models trained on simulations to real measurements mitigates the data-scarcity problem, which intensifies with model scale. This enables the development of new methods beyond current measurement capabilities and supports pre-training of large models on simulated data.