PP-MPI: A Deep Plug-and-Play Prior for Magnetic Particle Imaging Reconstruction

Askin, Baris; Güngör, Alper; Soydan, Damla Alptekin; Sarıtaş, Emine Ülkü; Top, Can Barış; Çukur, Tolga

doi:10.1007/978-3-031-17247-2_11

Cited by 11 publications

(12 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is observed that DIP shows relatively limited performance in this paper, and one possible reason is that its update only relies on the DC term (Dittmer et al 2020, Askin et al 2022. It might lead to a high sensitivity to noise and a less accurate reconstruction result, especially when dealing with complex imaging tasks or datasets with high noise levels.…”

Section: Discussionmentioning

confidence: 93%

“…PP-MPI applies a trained denoising network as the regularization term and integrates it into the ADMM algorithm. Like DIP, PP-MPI does not require specifying regularization terms and achieves much faster reconstruction speed (Askin et al 2022). However, directly adopting an image denoising model as a reconstruction prior may potentially limit its performance, and the iteration number remains undetermined.…”

Section: Deep Learning-based Methodsmentioning

confidence: 99%

“…and the optimal a TV was also chosen through experimentation. e was adjusted with SNRs, and it adopted the Euclidean distance of the noise (Ilbey et al 2017, Askin et al 2022). The optimal parameters were displayed in the table 1.…”

Section: Competing Methodsmentioning

confidence: 99%

“…(c) PP-MPI. The PP-MPI is deep-learning-based method, and it used a trained denoiser as the regularization terms (Askin et al 2022). For the training of the denoiser, LR was 0.01, and the standard deviation of the noise was 0.001. and 4000 iterations with ADAM optimizer were employed.…”

Section: Competing Methodsmentioning

confidence: 99%

“…(b) ADMM. We implemented ADMM algorithm with three regularization terms: L1, TV and L1+TV(Ilbey et al 2017, Askin et al 2022. The parameter m was optimized through experimentation and the number of iterations was 200.…”

mentioning

confidence: 99%

See 4 more Smart Citations

DERnet: a deep neural network for end-to-end reconstruction in magnetic particle imaging

Peng,

Yin,

Sun

et al. 2023

Phys. Med. Biol.

View full text Add to dashboard Cite

Objective. Magnetic particle imaging (MPI) shows potential for contributing to biomedical research and clinical practice. However, MPI images are effectively affected by noise in the signal as its reconstruction is an ill-posed inverse problem. Thus, effective reconstruction method is required to reduce the impact of the noise while mapping signals to MPI images. Traditional methods rely on the hand-crafted data-consistency (DC) term and regularization term based on spatial priors to achieve noise-reducing and reconstruction. While these methods alleviate the ill-posedness and reduce noise effects, they may be difficult to fully capture spatial features. Approach. In this study, we propose a deep neural network for end-to-end reconstruction (DERnet) in MPI that emulates the DC term and regularization term using the feature mapping subnetwork and post-processing subnetwork, respectively, but in a data-driven manner. By doing so, DERnet can better capture signal and spatial features without relying on hand-crafted priors and strategies, thereby effectively reducing noise interference and achieving superior reconstruction quality. Main results. Our data-driven method outperforms the state-of-the-art algorithms with an improvement of 0.9–8.8 dB in terms of peak signal-to-noise ratio under various noise levels. The result demonstrates the advantages of our approach in suppressing noise interference. Furthermore, DERnet can be employed for measured data reconstruction with improved fidelity and reduced noise. In conclusion, our proposed method offers performance benefits in reducing noise interference and enhancing reconstruction quality by effectively capturing signal and spatial features. Significance. DERnet is a promising candidate method to improve MPI reconstruction performance and facilitate its more in-depth biomedical application.

show abstract

Section: Discussionmentioning

confidence: 93%

Section: Deep Learning-based Methodsmentioning

confidence: 99%

Section: Competing Methodsmentioning

confidence: 99%

Section: Competing Methodsmentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

DERnet: a deep neural network for end-to-end reconstruction in magnetic particle imaging

Peng,

Yin,

Sun

et al. 2023

Phys. Med. Biol.

View full text Add to dashboard Cite

show abstract

Machine Learning and Deep Learning Applications in Magnetic Particle Imaging

Nigam,

Gjelaj,

Wang

et al. 2024

Magnetic Resonance Imaging

View full text Add to dashboard Cite

In recent years, magnetic particle imaging (MPI) has emerged as a promising imaging technique depicting high sensitivity and spatial resolution. It originated in the early 2000s where it proposed a new approach to challenge the low spatial resolution achieved by using relaxometry in order to measure the magnetic fields. MPI presents 2D and 3D images with high temporal resolution, non‐ionizing radiation, and optimal visual contrast due to its lack of background tissue signal. Traditionally, the images were reconstructed by the conversion of signal from the induced voltage by generating system matrix and X‐space based methods. Because image reconstruction and analyses play an integral role in obtaining precise information from MPI signals, newer artificial intelligence‐based methods are continuously being researched and developed upon. In this work, we summarize and review the significance and employment of machine learning and deep learning models for applications with MPI and the potential they hold for the future.Level of Evidence5Technical EfficacyStage 1

show abstract

MPIGAN: An end‐to‐end deep based generative framework for high‐resolution magnetic particle imaging reconstruction

Zhao,

Shen,

Liu

et al. 2024

Medical Physics

View full text Add to dashboard Cite

BackgroundMagnetic particle imaging (MPI) is a recently developed, non‐invasive in vivo imaging technique to map the spatial distribution of superparamagnetic iron oxide nanoparticles (SPIONs) in animal tissues with high sensitivity and speed. It is a challenge to reconstruct images directly from the received signals of MPI device due to the complex physical behavior of the nanoparticles. System matrix and X‐space are two commonly used MPI reconstruction methods, where the former is extremely time‐consuming and the latter usually produces blurry images.PurposeCurrently, we proposed an end‐to‐end machine learning framework to reconstruct high‐resolution MPI images from 1‐D voltage signals directly and efficiently.MethodsThe proposed framework, which we termed “MPIGAN”, was trained on a large MPI simulation dataset containing 291 597 pairs of high‐resolution 2‐D phantom images and each image's corresponding voltage signals, so that it was able to accurately capture the nonlinear relationship between the spatial distribution of SPIONs and the received voltage signal, and realized high‐resolution MPI image reconstruction.ResultsExperiment results showed that, MPIGAN exhibited remarkable abilities in high‐resolution MPI image reconstruction. MPIGAN outperformed the traditional methods of system matrix and X‐space in recovering the fine‐scale structure of magnetic nanoparticles’ spatial distribution and achieving enhanced reconstruction performance in both visual effects and quantitative assessments. Moreover, even when the received signals were severely contaminated with noise, MPIGAN could still generate high‐quality MPI images.ConclusionOur study provides a promising AI solution for end‐to‐end, efficient, and high‐resolution magnetic particle imaging reconstruction.

show abstract

PP-MPI: A Deep Plug-and-Play Prior for Magnetic Particle Imaging Reconstruction

Cited by 11 publications

References 13 publications

DERnet: a deep neural network for end-to-end reconstruction in magnetic particle imaging

DERnet: a deep neural network for end-to-end reconstruction in magnetic particle imaging

Machine Learning and Deep Learning Applications in Magnetic Particle Imaging

MPIGAN: An end‐to‐end deep based generative framework for high‐resolution magnetic particle imaging reconstruction

Contact Info

Product

Resources

About