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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep learning (DL) models are capable of successfully exploiting latentrepresentations in MR data and have become state-of-the-art for accelerated MRIreconstruction. However, undersampling the measurements in k-space as well asthe over- or under-parameterized and non-transparent nature of DL make thesemodels exposed to uncertainty. Consequently, uncertainty estimation has becomea major issue in DL MRI reconstruction. To estimate uncertainty, Monte Carlo(MC) inference techniques have become a common practice where multiplereconstructions are utilized to compute the variance in reconstruction as ameasurement of uncertainty. However, these methods demand high computationalcosts as they require multiple inferences through the DL model. To this end, weintroduce a method to estimate uncertainty during MRI reconstruction using apixel classification framework. The proposed method, PixCUE (stands for PixelClassification Uncertainty Estimation) produces the reconstructed image alongwith an uncertainty map during a single forward pass through the DL model. Wedemonstrate that this approach generates uncertainty maps that highly correlatewith the reconstruction errors with respect to various MR imaging sequences andunder numerous adversarial conditions. We also show that the estimateduncertainties are correlated to that of the conventional MC method. We furtherprovide an empirical relationship between the uncertainty estimations usingPixCUE and well-established reconstruction metrics such as NMSE, PSNR, andSSIM. We conclude that PixCUE is capable of reliably estimating the uncertaintyin MRI reconstruction with a minimum additional computational cost.

PixCUE: Joint Uncertainty Estimation and Image Reconstruction in MRI using Deep Pixel Classification