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

Recently, joint registration and segmentation has been formulated in a deeplearning setting, by the definition of joint loss functions. In this work, weinvestigate joining these tasks at the architectural level. We propose aregistration network that integrates segmentation propagation between images,and a segmentation network to predict the segmentation directly. These networksare connected into a single joint architecture via so-called cross-stitchunits, allowing information to be exchanged between the tasks in a learnablemanner. The proposed method is evaluated in the context of adaptiveimage-guided radiotherapy, using daily prostate CT imaging. Two datasets fromdifferent institutes and manufacturers were involved in the study. The firstdataset was used for training (12 patients) and validation (6 patients), whilethe second dataset was used as an independent test set (14 patients). In termsof mean surface distance, our approach achieved $1.06 \pm 0.3$ mm, $0.91 \pm0.4$ mm, $1.27 \pm 0.4$ mm, and $1.76 \pm 0.8$ mm on the validation set and$1.82 \pm 2.4$ mm, $2.45 \pm 2.4$ mm, $2.45 \pm 5.0$ mm, and $2.57 \pm 2.3$ mmon the test set for the prostate, bladder, seminal vesicles, and rectum,respectively. The proposed multi-task network outperformed single-tasknetworks, as well as a network only joined through the loss function, thusdemonstrating the capability to leverage the individual strengths of thesegmentation and registration tasks. The obtained performance as well as theinference speed make this a promising candidate for daily re-contouring inadaptive radiotherapy, potentially reducing treatment-related side effects andimproving quality-of-life after treatment.

A Cross-Stitch Architecture for Joint Registration and Segmentation in Adaptive Radiotherapy