International Journal on Magnetic Particle Imaging IJMPI
Vol. 12 No. 1 Suppl 1 (2026): Int J Mag Part Imag
https://doi.org/10.18416/IJMPI.2026.2603037
Proceedings Articles, ID 990
Deep image prior for alternating direction method of multipliers reconstruction of experimental magnetic particle imaging data
Main Article Content
Copyright (c) 2026 Antonios Nikolakis-Plytzanopoulos, George A. Kastis, Jing Zhong, Nikolaos Dikaios
This work is licensed under a Creative Commons Attribution 4.0 International License.
Abstract
This work presents an on-the-fly training Deep Image Prior (DIP) integrated into an Alternating Direction Method of Multipliers (ADMM) algorithm for Magnetic Particle Imaging (MPI) system-matrix reconstruction. Experimental data results indicate competitive quality and robustness, supporting the approach as a practical alternative to classical Tikhonov.
Article Details
References
linear response of magnetic particles. Nature, 435(7046):1214–1217,
2005, doi:10.1038/nature03808.
[2] L. Yin, W. Li, Y. Du, K. Wang, Z. Liu, H. Hui, and J. Tian. Recent
developments of the reconstruction in magnetic particle imaging.
Visual computing for industry, biomedicine, and art, 5(1):24, 2022.
[3] S. Dittmer, T. Kluth, M. T. R. Henriksen, and P. Maass, Deep image
prior for 3d magnetic particle imaging: A quantitative compari-
son of regularization techniques on open mpi dataset, 2020. arXiv:
2007.01593.
[4] S. Ilbey, C. B. Top, A. Güngör, T. Çukur, E. U. Saritas, and H. E.
Güven. Comparison of system-matrix-based and projection-based
reconstructions for field free line magnetic particle imaging. Inter-
national Journal on Magnetic Particle Imaging IJMPI, 3(1), 2017.
[5] R. Zhang, S. Sun, S. Sun, M. Meribout, L. Xu, and J. Zhong.
Mixing-space magnetic particle imaging. IEEE Transac-
tions on Instrumentation and Measurement, 74:1–9, 2025,
doi:10.1109/TIM.2024.3500075.