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

Limited capture range, and the requirement to provide high qualityinitialization for optimization-based 2D/3D image registration methods, cansignificantly degrade the performance of 3D image reconstruction and motioncompensation pipelines. Challenging clinical imaging scenarios, which containsignificant subject motion such as fetal in-utero imaging, complicate the 3Dimage and volume reconstruction process. In this paper we present a learningbased image registration method capable of predicting 3D rigid transformationsof arbitrarily oriented 2D image slices, with respect to a learned canonicalatlas co-ordinate system. Only image slice intensity information is used toperform registration and canonical alignment, no spatial transforminitialization is required. To find image transformations we utilize aConvolutional Neural Network (CNN) architecture to learn the regressionfunction capable of mapping 2D image slices to a 3D canonical atlas space. Weextensively evaluate the effectiveness of our approach quantitatively onsimulated Magnetic Resonance Imaging (MRI), fetal brain imagery with syntheticmotion and further demonstrate qualitative results on real fetal MRI data whereour method is integrated into a full reconstruction and motion compensationpipeline. Our learning based registration achieves an average spatialprediction error of 7 mm on simulated data and produces qualitatively improvedreconstructions for heavily moving fetuses with gestational ages ofapproximately 20 weeks. Our model provides a general and computationallyefficient solution to the 2D/3D registration initialization problem and issuitable for real-time scenarios.

3D Reconstruction in Canonical Co-ordinate Space from Arbitrarily Oriented 2D Images