English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Quantitative MRI (qMRI) aims to map tissue properties non-invasively viamodels that relate these unknown quantities to measured MRI signals. Estimatingthese unknowns, which has traditionally required model fitting - an ofteniterative procedure, can now be done with one-shot machine learning (ML)approaches. Such parameter estimation may be complicated by intrinsic qMRIsignal model degeneracy: different combinations of tissue properties producethe same signal. Despite their many advantages, it remains unclear whether MLapproaches can resolve this issue. Growing empirical evidence appears tosuggest ML approaches remain susceptible to model degeneracy. Here wedemonstrate under the right circumstances ML can address this issue. Inspiredby recent works on the impact of training data distributions on ML-basedparameter estimation, we propose to resolve model degeneracy by designingtraining data distributions. We put forward a classification of modeldegeneracies and identify one particular kind of degeneracies amenable to theproposed attack. The strategy is demonstrated successfully using the RevisedNODDI model with standard multi-shell diffusion MRI data as an exemplar. Ourresults illustrate the importance of training set design which has thepotential to allow accurate estimation of tissue properties with ML.

Resolving quantitative MRI model degeneracy with machine learning via training data distribution design