An $\ell^1$-Plug-and-Play Approach for MPI Using a Zero Shot Denoiser with Evaluation on the 3D Open MPI Dataset
Authors:
Vladyslav Gapyak,
Corinna Rentschler,
Thomas März,
Andreas Weinmann
Abstract:
Objective: Magnetic particle imaging (MPI) is an emerging medical imaging modality which has gained increasing interest in recent years. Among the benefits of MPI are its high temporal resolution, and that the technique does not expose the specimen to any kind of ionizing radiation. It is based on the non-linear response of magnetic nanoparticles to an applied magnetic field. From the electric sig…
▽ More
Objective: Magnetic particle imaging (MPI) is an emerging medical imaging modality which has gained increasing interest in recent years. Among the benefits of MPI are its high temporal resolution, and that the technique does not expose the specimen to any kind of ionizing radiation. It is based on the non-linear response of magnetic nanoparticles to an applied magnetic field. From the electric signal measured in receive coils, the particle concentration has to be reconstructed. Due to the ill-posedness of the reconstruction problem, various regularization methods have been proposed for reconstruction ranging from early stop** methods, via classical Tikhonov regularization and iterative methods to modern machine learning approaches. In this work, we contribute to the latter class: we propose a plug-and-play approach based on a generic zero-shot denoiser with an $\ell^1$-prior.
Approach: We validate the reconstruction parameters of the method on a hybrid dataset and compare it with the baseline Tikhonov, DIP and the previous PP-MPI, which is a plug-and-play method with denoiser trained on MPI-friendly data.
Main results: We offer a quantitative and qualitative evaluation of the zero-shot plug-and-play approach on the 3D Open MPI dataset. Moreover, we show the quality of the approach with different levels of preprocessing of the data.
Significance: The proposed method employs a zero-shot denoiser which has not been trained for the MPI task and therefore saves the cost for training. Moreover, it offers a method that can be potentially applied in future MPI contexts.
△ Less
Submitted 22 May, 2024; v1 submitted 30 December, 2023;
originally announced January 2024.
Highly efficient narrowband terahertz generation driven by a two-spectral-line laser in PPLN
Authors:
H. T. Olgun,
W. Tian,
G. Cirmi,
K. Ravi,
C. Rentschler,
H. Çankaya,
M. Pergament,
M. Hemmer,
N. H. Matlis,
F. X. Kärtner
Abstract:
We demonstrate record conversion efficiencies of up to 0.9 percent for narrowband (1 percent relative bandwidth) THz generation by strongly-cascaded difference frequency generation. These results are achieved using a novel laser source, customized for high efficiencies, with two narrow spectral lines of variable separation and pulse duration (250 ps). THz generation in 5 percent MgO doped PPLN cry…
▽ More
We demonstrate record conversion efficiencies of up to 0.9 percent for narrowband (1 percent relative bandwidth) THz generation by strongly-cascaded difference frequency generation. These results are achieved using a novel laser source, customized for high efficiencies, with two narrow spectral lines of variable separation and pulse duration (250 ps). THz generation in 5 percent MgO doped PPLN crystals of varying poling period was explored at cryogenic and room temperature operation as well as different crystal lengths. This work addresses an increasing demand for high-field THz pulses which has, up to now, been largely limited by low optical-to-THz conversion efficiencies.
△ Less
Submitted 11 November, 2021;
originally announced November 2021.
