FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3201607300> ?p ?o ?g. }

• W3201607300 abstract "Neural networks are increasingly used to estimate parameters in quantitative MRI, in particular in magnetic resonance fingerprinting. Their advantages over the gold standard non-linear least square fitting are their superior speed and their immunity to the non-convexity of many fitting problems. We find, however, that in heterogeneous parameter spaces, i.e. in spaces in which the variance of the estimated parameters varies considerably, good performance is hard to achieve and requires arduous tweaking of the loss function, hyper parameters, and the distribution of the training data in parameter space. Here, we address these issues with a theoretically well-founded loss function: the Cram'er-Rao bound (CRB) provides a theoretical lower bound for the variance of an unbiased estimator and we propose to normalize the squared error with respective CRB. With this normalization, we balance the contributions of hard-to-estimate and not-so-hard-to-estimate parameters and areas in parameter space, and avoid a dominance of the former in the overall training loss. Further, the CRB-based loss function equals one for a maximally-efficient unbiased estimator, which we consider the ideal estimator. Hence, the proposed CRB-based loss function provides an absolute evaluation metric. We compare a network trained with the CRB-based loss with a network trained with the commonly used means squared error loss and demonstrate the advantages of the former in numerical, phantom, and in vivo experiments." @default.
• W3201607300 created "2021-09-27" @default.
• W3201607300 creator A5034482431 @default.
• W3201607300 creator A5034694275 @default.
• W3201607300 creator A5044336556 @default.
• W3201607300 creator A5065517740 @default.
• W3201607300 creator A5080050038 @default.
• W3201607300 creator A5086328508 @default.
• W3201607300 creator A5086517621 @default.
• W3201607300 date "2021-09-22" @default.
• W3201607300 modified "2023-09-27" @default.
• W3201607300 title "Cramér-Rao bound-informed training of neural networks for quantitative MRI." @default.
• W3201607300 cites W1479979375 @default.
• W3201607300 cites W1522301498 @default.
• W3201607300 cites W1867572349 @default.
• W3201607300 cites W1980940294 @default.
• W3201607300 cites W1987611978 @default.
• W3201607300 cites W2009524946 @default.
• W3201607300 cites W2029816571 @default.
• W3201607300 cites W2035808554 @default.
• W3201607300 cites W2046748764 @default.
• W3201607300 cites W2061708033 @default.
• W3201607300 cites W2107915383 @default.
• W3201607300 cites W2144288697 @default.
• W3201607300 cites W2151354228 @default.
• W3201607300 cites W2167227980 @default.
• W3201607300 cites W2175306913 @default.
• W3201607300 cites W2177835590 @default.
• W3201607300 cites W2194775991 @default.
• W3201607300 cites W2535510318 @default.
• W3201607300 cites W2747122024 @default.
• W3201607300 cites W2792643794 @default.
• W3201607300 cites W2795783309 @default.
• W3201607300 cites W2798456213 @default.
• W3201607300 cites W2902147031 @default.
• W3201607300 cites W2914314139 @default.
• W3201607300 cites W2937047179 @default.
• W3201607300 cites W2950000029 @default.
• W3201607300 cites W2962689831 @default.
• W3201607300 cites W2962996460 @default.
• W3201607300 cites W2967215079 @default.
• W3201607300 cites W2972976643 @default.
• W3201607300 cites W2994844425 @default.
• W3201607300 cites W3015961530 @default.
• W3201607300 cites W3048541905 @default.
• W3201607300 cites W3099449442 @default.
• W3201607300 cites W585064526 @default.
• W3201607300 cites W2168887049 @default.
• W3201607300 hasPublicationYear "2021" @default.
• W3201607300 type Work @default.
• W3201607300 sameAs 3201607300 @default.
• W3201607300 citedByCount "0" @default.
• W3201607300 crossrefType "posted-content" @default.
• W3201607300 hasAuthorship W3201607300A5034482431 @default.
• W3201607300 hasAuthorship W3201607300A5034694275 @default.
• W3201607300 hasAuthorship W3201607300A5044336556 @default.
• W3201607300 hasAuthorship W3201607300A5065517740 @default.
• W3201607300 hasAuthorship W3201607300A5080050038 @default.
• W3201607300 hasAuthorship W3201607300A5086328508 @default.
• W3201607300 hasAuthorship W3201607300A5086517621 @default.
• W3201607300 hasConcept C105795698 @default.
• W3201607300 hasConcept C11413529 @default.
• W3201607300 hasConcept C121955636 @default.
• W3201607300 hasConcept C126255220 @default.
• W3201607300 hasConcept C136886441 @default.
• W3201607300 hasConcept C139945424 @default.
• W3201607300 hasConcept C144024400 @default.
• W3201607300 hasConcept C144133560 @default.
• W3201607300 hasConcept C154945302 @default.
• W3201607300 hasConcept C165646398 @default.
• W3201607300 hasConcept C185429906 @default.
• W3201607300 hasConcept C191393472 @default.
• W3201607300 hasConcept C19165224 @default.
• W3201607300 hasConcept C196083921 @default.
• W3201607300 hasConcept C28826006 @default.
• W3201607300 hasConcept C33923547 @default.
• W3201607300 hasConcept C41008148 @default.
• W3201607300 hasConcept C4978587 @default.
• W3201607300 hasConcept C50644808 @default.
• W3201607300 hasConcept C73586568 @default.
• W3201607300 hasConceptScore W3201607300C105795698 @default.
• W3201607300 hasConceptScore W3201607300C11413529 @default.
• W3201607300 hasConceptScore W3201607300C121955636 @default.
• W3201607300 hasConceptScore W3201607300C126255220 @default.
• W3201607300 hasConceptScore W3201607300C136886441 @default.
• W3201607300 hasConceptScore W3201607300C139945424 @default.
• W3201607300 hasConceptScore W3201607300C144024400 @default.
• W3201607300 hasConceptScore W3201607300C144133560 @default.
• W3201607300 hasConceptScore W3201607300C154945302 @default.
• W3201607300 hasConceptScore W3201607300C165646398 @default.
• W3201607300 hasConceptScore W3201607300C185429906 @default.
• W3201607300 hasConceptScore W3201607300C191393472 @default.
• W3201607300 hasConceptScore W3201607300C19165224 @default.
• W3201607300 hasConceptScore W3201607300C196083921 @default.
• W3201607300 hasConceptScore W3201607300C28826006 @default.
• W3201607300 hasConceptScore W3201607300C33923547 @default.
• W3201607300 hasConceptScore W3201607300C41008148 @default.
• W3201607300 hasConceptScore W3201607300C4978587 @default.
• W3201607300 hasConceptScore W3201607300C50644808 @default.
• W3201607300 hasConceptScore W3201607300C73586568 @default.

• next