Technology Category
Machine Learning: Imitation Learning & Inverse Reinforcement LearningComputer Vision: Medical and Biological ImagingSearch and Optimization: Learning to Search
Learned Discrepancy Reconstruction and Benchmark Dataset for Magnetic Particle Imaging
🤖 AI Summary
Non-Gaussian noise in magnetic particle imaging (MPI) severely degrades image reconstruction fidelity. Method: This paper proposes a diffractive learning reconstruction framework based on invertible neural networks (INNs), the first to embed INNs within an inverse problem formulation to explicitly model MPI-specific noise distributions. We further introduce MPI-MNIST—the first open-source, algorithm-evaluation-oriented simulation benchmark—integrating multiple noise sources, realistic system matrices, and experimental measurements from a preclinical scanner. Our approach adopts a hybrid model- and data-driven paradigm, combining an MNIST-guided MPI simulation pipeline with high-fidelity preclinical system modeling. Results: On MPI-MNIST, our method achieves significantly higher structural similarity (SSIM) than conventional methods including Tikhonov and L1-regularized reconstructions, demonstrating superior robustness to complex noise and strong generalization capability.
Technology Category
Application Category
📝 Abstract
Magnetic Particle Imaging (MPI) is an emerging imaging modality based on the magnetic response of superparamagnetic iron oxide nanoparticles to achieve high-resolution and real-time imaging without harmful radiation. One key challenge in the MPI image reconstruction task arises from its underlying noise model, which does not fulfill the implicit Gaussian assumptions that are made when applying traditional reconstruction approaches. To address this challenge, we introduce the Learned Discrepancy Approach, a novel learning-based reconstruction method for inverse problems that includes a learned discrepancy function. It enhances traditional techniques by incorporating an invertible neural network to explicitly model problem-specific noise distributions. This approach does not rely on implicit Gaussian noise assumptions, making it especially suited to handle the sophisticated noise model in MPI and also applicable to other inverse problems. To further advance MPI reconstruction techniques, we introduce the MPI-MNIST dataset - a large collection of simulated MPI measurements derived from the MNIST dataset of handwritten digits. The dataset includes noise-perturbed measurements generated from state-of-the-art model-based system matrices and measurements of a preclinical MPI scanner device. This provides a realistic and flexible environment for algorithm testing. Validated against the MPI-MNIST dataset, our method demonstrates significant improvements in reconstruction quality in terms of structural similarity when compared to classical reconstruction techniques.
Problem
Research questions and friction points this paper is trying to address.
Magnetic Particle Imaging
Noise Pattern
Image Reconstruction
Innovation
Methods, ideas, or system contributions that make the work stand out.
Learned Discrepancy Approach
Reversible Neural Networks
MPI-MNIST Dataset
🔎 Similar Papers
