FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4381157953> ?p ?o ?g. }

• W4381157953 endingPage "1761" @default.
• W4381157953 startingPage "1749" @default.
• W4381157953 abstract "The inherently poor SNR of MRS measurements presents a significant hurdle to its clinical application. Denoising by machine or deep learning (DL) was proposed as a remedy. It is investigated whether such denoising leads to lower estimate uncertainties or whether it essentially reduces noise in signal-free areas only.Noise removal based on supervised DL with U-nets was implemented using simulated 1 H MR spectra of human brain in two approaches: (1) via time-frequency domain spectrograms and (2) using 1D spectra as input. Quality of denoising was evaluated in three ways: (1) by an adapted fit quality score, (2) by traditional model fitting, and (3) by quantification via neural networks.Visually appealing spectra were obtained; hinting that denoising is well-suited for MRS. However, an adapted denoising score showed that noise removal is inhomogeneous and more efficient for signal-free areas. This was confirmed by quantitative analysis of traditional fit results as well as DL quantitation following DL denoising. DL denoising, although apparently successful as judged by mean squared errors, led to substantially biased estimates in both implementations.The implemented DL-based denoising techniques may be useful for display purposes, but do not help quantitative evaluations, confirming expectations based on estimation theory: Cramér Rao lower bounds defined by the original data and the appropriate fitting model cannot be circumvented in an unbiased way for single data sets, unless additional prior knowledge can be incurred in the form of parameter restrictions/relations or applicable substates." @default.
• W4381157953 created "2023-06-20" @default.
• W4381157953 creator A5052827424 @default.
• W4381157953 creator A5055218218 @default.
• W4381157953 creator A5059123000 @default.
• W4381157953 creator A5064005376 @default.
• W4381157953 date "2023-06-18" @default.
• W4381157953 modified "2023-10-01" @default.
• W4381157953 title "Denoising single <scp>MR</scp> spectra by deep learning: Miracle or mirage?" @default.
• W4381157953 cites W160168085 @default.
• W4381157953 cites W2005615201 @default.
• W4381157953 cites W2035068046 @default.
• W4381157953 cites W2050344398 @default.
• W4381157953 cites W2058453951 @default.
• W4381157953 cites W2083710416 @default.
• W4381157953 cites W2094597486 @default.
• W4381157953 cites W2104058133 @default.
• W4381157953 cites W2108469347 @default.
• W4381157953 cites W2158514967 @default.
• W4381157953 cites W2168936817 @default.
• W4381157953 cites W2492985850 @default.
• W4381157953 cites W2518702244 @default.
• W4381157953 cites W2610163897 @default.
• W4381157953 cites W2726456930 @default.
• W4381157953 cites W2754059241 @default.
• W4381157953 cites W2761128463 @default.
• W4381157953 cites W2788193639 @default.
• W4381157953 cites W2790628621 @default.
• W4381157953 cites W2800950215 @default.
• W4381157953 cites W2887890081 @default.
• W4381157953 cites W2912052020 @default.
• W4381157953 cites W2918344699 @default.
• W4381157953 cites W2921982517 @default.
• W4381157953 cites W2963091230 @default.
• W4381157953 cites W2967528850 @default.
• W4381157953 cites W2970506711 @default.
• W4381157953 cites W3009564973 @default.
• W4381157953 cites W3017296936 @default.
• W4381157953 cites W3017944199 @default.
• W4381157953 cites W3034926329 @default.
• W4381157953 cites W3080415167 @default.
• W4381157953 cites W3128745853 @default.
• W4381157953 cites W3133019246 @default.
• W4381157953 cites W3163627798 @default.
• W4381157953 cites W3164762844 @default.
• W4381157953 cites W3201041228 @default.
• W4381157953 cites W4296119719 @default.
• W4381157953 cites W4296666707 @default.
• W4381157953 cites W4311934431 @default.
• W4381157953 cites W4362602739 @default.
• W4381157953 cites W3155413364 @default.
• W4381157953 doi "https://doi.org/10.1002/mrm.29762" @default.
• W4381157953 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37332185" @default.
• W4381157953 hasPublicationYear "2023" @default.
• W4381157953 type Work @default.
• W4381157953 citedByCount "2" @default.
• W4381157953 countsByYear W43811579532023 @default.
• W4381157953 crossrefType "journal-article" @default.
• W4381157953 hasAuthorship W4381157953A5052827424 @default.
• W4381157953 hasAuthorship W4381157953A5055218218 @default.
• W4381157953 hasAuthorship W4381157953A5059123000 @default.
• W4381157953 hasAuthorship W4381157953A5064005376 @default.
• W4381157953 hasBestOaLocation W43811579531 @default.
• W4381157953 hasConcept C11413529 @default.
• W4381157953 hasConcept C115961682 @default.
• W4381157953 hasConcept C119857082 @default.
• W4381157953 hasConcept C13944312 @default.
• W4381157953 hasConcept C153180895 @default.
• W4381157953 hasConcept C154945302 @default.
• W4381157953 hasConcept C163294075 @default.
• W4381157953 hasConcept C199360897 @default.
• W4381157953 hasConcept C2779843651 @default.
• W4381157953 hasConcept C41008148 @default.
• W4381157953 hasConcept C45273575 @default.
• W4381157953 hasConcept C50644808 @default.
• W4381157953 hasConcept C76155785 @default.
• W4381157953 hasConcept C99498987 @default.
• W4381157953 hasConceptScore W4381157953C11413529 @default.
• W4381157953 hasConceptScore W4381157953C115961682 @default.
• W4381157953 hasConceptScore W4381157953C119857082 @default.
• W4381157953 hasConceptScore W4381157953C13944312 @default.
• W4381157953 hasConceptScore W4381157953C153180895 @default.
• W4381157953 hasConceptScore W4381157953C154945302 @default.
• W4381157953 hasConceptScore W4381157953C163294075 @default.
• W4381157953 hasConceptScore W4381157953C199360897 @default.
• W4381157953 hasConceptScore W4381157953C2779843651 @default.
• W4381157953 hasConceptScore W4381157953C41008148 @default.
• W4381157953 hasConceptScore W4381157953C45273575 @default.
• W4381157953 hasConceptScore W4381157953C50644808 @default.
• W4381157953 hasConceptScore W4381157953C76155785 @default.
• W4381157953 hasConceptScore W4381157953C99498987 @default.
• W4381157953 hasFunder F4320320924 @default.
• W4381157953 hasFunder F4320338337 @default.
• W4381157953 hasIssue "5" @default.
• W4381157953 hasLocation W43811579531 @default.
• W4381157953 hasLocation W43811579532 @default.
• W4381157953 hasLocation W43811579533 @default.
• W4381157953 hasOpenAccess W4381157953 @default.

• next