MAGORINO : Magnitude‐only fat fraction and R 2 estimation with Rician noise modeling

Magnitude-based fitting of chemical shift-encoded data enables proton density fat fraction (PDFF) and estimation where complex-based methods fail or when phase data are inaccessible or unreliable. However, traditional magnitude-based fitting algorithms do not account for Rician noise, creating a sou...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Magnetic resonance in medicine Ročník 89; číslo 3; s. 1173 - 1192
Hlavní autoři: Bray, Timothy J. P., Bainbridge, Alan, Lim, Emma, Hall‐Craggs, Margaret A., Zhang, Hui
Médium: Journal Article
Jazyk:angličtina
Vydáno: United States John Wiley and Sons Inc 01.03.2023
Témata:
Adipose Tissue - diagnostic imaging
Algorithms
Computer Simulation
Likelihood Functions
Magnetic Resonance Imaging - methods
Normal Distribution
Protons
Reproducibility of Results
s—Computer Processing and Modeling
computer-assisted imaging processing
magnetic resonance imaging
radiology
ISSN:0740-3194, 1522-2594, 1522-2594
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Magnitude-based fitting of chemical shift-encoded data enables proton density fat fraction (PDFF) and estimation where complex-based methods fail or when phase data are inaccessible or unreliable. However, traditional magnitude-based fitting algorithms do not account for Rician noise, creating a source of bias. To address these issues, we propose an algorithm for magnitude-only PDFF and estimation with Rician noise modeling (MAGORINO). Simulations of multi-echo gradient-echo signal intensities are used to investigate the performance and behavior of MAGORINO over the space of clinically plausible PDFF, , and SNR values. Fitting performance is assessed through detailed simulation, including likelihood function visualization, and in a multisite, multivendor, and multi-field-strength phantom data set and in vivo. Simulations show that Rician noise-based magnitude fitting outperforms existing Gaussian noise-based fitting and reveals two key mechanisms underpinning the observed improvement. First, the likelihood functions exhibit two local optima; Rician noise modeling increases the chance that the global optimum corresponds to the ground truth. Second, when the global optimum corresponds to ground truth for both noise models, the optimum from Rician noise modeling is closer to ground truth. Multisite phantom experiments show good agreement of MAGORINO PDFF with reference values, and in vivo experiments replicate the performance benefits observed in simulation. The MAGORINO algorithm reduces Rician noise-related bias in PDFF and estimation, thus addressing a key limitation of existing magnitude-only fitting methods. Our results offer insight into the importance of the noise model for selecting the correct optimum when multiple plausible optima exist.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
Funding information Research Trainees Coordinating Centre, Grant/Award Number: CL‐2019‐18‐001; National Institute for Health Research, NIHR, UCLH Biomedical Research Centre
ISSN:0740-3194
1522-2594
1522-2594
DOI:10.1002/mrm.29493