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 based computed tomography (CT) reconstruction has demonstratedoutstanding performance on simulated 2D low-dose CT data. This applies inparticular to domain adapted neural networks, which incorporate a handcraftedphysics model for CT imaging. Empirical evidence shows that employing sucharchitectures reduces the demand for training data and improves upongeneralisation. However, their training requires large computational resourcesthat quickly become prohibitive in 3D helical CT, which is the most commonacquisition geometry used for medical imaging. Furthermore, clinical data alsocomes with other challenges not accounted for in simulations, like errors influx measurement, resolution mismatch and, most importantly, the absence of thereal ground truth. The necessity to have a computationally feasible trainingcombined with the need to address these issues has made it difficult toevaluate deep learning based reconstruction on clinical 3D helical CT. Thispaper modifies a domain adapted neural network architecture, the LearnedPrimal-Dual (LPD), so that it can be trained and applied to reconstruction inthis setting. We achieve this by splitting the helical trajectory into sectionsand applying the unrolled LPD iterations to those sections sequentially. To thebest of our knowledge, this work is the first to apply an unrolled deeplearning architecture for reconstruction on full-sized clinical data, likethose in the Low dose CT image and projection data set (LDCT). Moreover,training and testing is done on a single GPU card with 24GB of memory.

3D helical CT Reconstruction with a Memory Efficient Learned Primal-Dual Architecture