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/

Zendy Plus

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

Zendy Tools

Zendy Open

Diffusion MRI (dMRI) of the developing brain can provide valuable insightsinto the white matter development. However, slice thickness in fetal dMRI istypically high (i.e., 3-5 mm) to freeze the in-plane motion, which reduces thesensitivity of the dMRI signal to the underlying anatomy. In this study, we aimat overcoming this problem by using autoencoders to learn unsupervisedefficient representations of brain slices in a latent space, using raw dMRIsignals and their spherical harmonics (SH) representation. We first learn andquantitatively validate the autoencoders on the developing Human ConnectomeProject pre-term newborn data, and further test the method on fetal data. Ourresults show that the autoencoder in the signal domain better synthesized theraw signal. Interestingly, the fractional anisotropy and, to a lesser extent,the mean diffusivity, are best recovered in missing slices by using theautoencoder trained with SH coefficients. A comparison was performed with thesame maps reconstructed using an autoencoder trained with raw signals, as wellas conventional interpolation methods of raw signals and SH coefficients. Fromthese results, we conclude that the recovery of missing/corrupted slices shouldbe performed in the signal domain if the raw signal is aimed to be recovered,and in the SH domain if diffusion tensor properties (i.e., fractionalanisotropy) are targeted. Notably, the trained autoencoders were able togeneralize to fetal dMRI data acquired using a much smaller number of diffusiongradients and a lower b-value, where we qualitatively show the consistency ofthe estimated diffusion tensor maps.

Slice estimation in diffusion MRI of neonatal and fetal brains in image and spherical harmonics domains using autoencoders