{"title":"Automated Parameter Selection for Accelerated MRI Reconstruction via Low-Rank Modeling of Local k-Space Neighborhoods","authors":"Efe Ilicak , Emine Ulku Saritas , Tolga Çukur","doi":"10.1016/j.zemedi.2022.02.002","DOIUrl":null,"url":null,"abstract":"<div><h3>Purpose</h3><p>Image quality in accelerated MRI rests on careful selection of various reconstruction parameters. A common yet tedious and error-prone practice is to hand-tune each parameter to attain visually appealing reconstructions. Here, we propose a parameter tuning strategy to automate hybrid parallel imaging (PI) – compressed sensing (CS) reconstructions via low-rank modeling of local k-space neighborhoods (LORAKS) supplemented with sparsity regularization in wavelet and total variation (TV) domains.</p></div><div><h3>Methods</h3><p>For low-rank regularization, we leverage a soft-thresholding operation based on singular values for matrix rank selection in LORAKS. For sparsity regularization, we employ Stein's unbiased risk estimate criterion to select the wavelet regularization parameter and local standard deviation of reconstructions to select the TV regularization parameter. Comprehensive demonstrations are presented on a numerical brain phantom and in vivo brain and knee acquisitions. Quantitative assessments are performed via PSNR, SSIM and NMSE metrics.</p></div><div><h3>Results</h3><p>The proposed hybrid PI-CS method improves reconstruction quality compared to PI-only techniques, and it achieves on par image quality to reconstructions with brute-force optimization of reconstruction parameters. These results are prominent across several different datasets and the range of examined acceleration rates.</p></div><div><h3>Conclusion</h3><p>A data-driven parameter tuning strategy to automate hybrid PI-CS reconstructions is presented. The proposed method achieves reliable reconstructions of accelerated multi-coil MRI datasets without the need for exhaustive hand-tuning of reconstruction parameters.</p></div>","PeriodicalId":54397,"journal":{"name":"Zeitschrift fur Medizinische Physik","volume":null,"pages":null},"PeriodicalIF":2.4000,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/8a/60/main.PMC10311279.pdf","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Zeitschrift fur Medizinische Physik","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0939388922000083","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}
引用次数: 3
Abstract
Purpose
Image quality in accelerated MRI rests on careful selection of various reconstruction parameters. A common yet tedious and error-prone practice is to hand-tune each parameter to attain visually appealing reconstructions. Here, we propose a parameter tuning strategy to automate hybrid parallel imaging (PI) – compressed sensing (CS) reconstructions via low-rank modeling of local k-space neighborhoods (LORAKS) supplemented with sparsity regularization in wavelet and total variation (TV) domains.
Methods
For low-rank regularization, we leverage a soft-thresholding operation based on singular values for matrix rank selection in LORAKS. For sparsity regularization, we employ Stein's unbiased risk estimate criterion to select the wavelet regularization parameter and local standard deviation of reconstructions to select the TV regularization parameter. Comprehensive demonstrations are presented on a numerical brain phantom and in vivo brain and knee acquisitions. Quantitative assessments are performed via PSNR, SSIM and NMSE metrics.
Results
The proposed hybrid PI-CS method improves reconstruction quality compared to PI-only techniques, and it achieves on par image quality to reconstructions with brute-force optimization of reconstruction parameters. These results are prominent across several different datasets and the range of examined acceleration rates.
Conclusion
A data-driven parameter tuning strategy to automate hybrid PI-CS reconstructions is presented. The proposed method achieves reliable reconstructions of accelerated multi-coil MRI datasets without the need for exhaustive hand-tuning of reconstruction parameters.
期刊介绍:
Zeitschrift fur Medizinische Physik (Journal of Medical Physics) is an official organ of the German and Austrian Society of Medical Physic and the Swiss Society of Radiobiology and Medical Physics.The Journal is a platform for basic research and practical applications of physical procedures in medical diagnostics and therapy. The articles are reviewed following international standards of peer reviewing.
Focuses of the articles are:
-Biophysical methods in radiation therapy and nuclear medicine
-Dosimetry and radiation protection
-Radiological diagnostics and quality assurance
-Modern imaging techniques, such as computed tomography, magnetic resonance imaging, positron emission tomography
-Ultrasonography diagnostics, application of laser and UV rays
-Electronic processing of biosignals
-Artificial intelligence and machine learning in medical physics
In the Journal, the latest scientific insights find their expression in the form of original articles, reviews, technical communications, and information for the clinical practice.