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

In oncology research, accurate 3D segmentation of lesions from CT scans isessential for the modeling of lesion growth kinetics. However, following theRECIST criteria, radiologists routinely only delineate each lesion on the axialslice showing the largest transverse area, and delineate a small number oflesions in 3D for research purposes. As a result, we have plenty of unlabeled3D volumes and labeled 2D images, and scarce labeled 3D volumes, which makestraining a deep-learning 3D segmentation model a challenging task. In thiswork, we propose a novel model, denoted a multi-dimension unified Swintransformer (MDU-ST), for 3D lesion segmentation. The MDU-ST consists of aShifted-window transformer (Swin-transformer) encoder and a convolutionalneural network (CNN) decoder, allowing it to adapt to 2D and 3D inputs andlearn the corresponding semantic information in the same encoder. Based on thismodel, we introduce a three-stage framework: 1) leveraging large amount ofunlabeled 3D lesion volumes through self-supervised pretext tasks to learn theunderlying pattern of lesion anatomy in the Swin-transformer encoder; 2)fine-tune the Swin-transformer encoder to perform 2D lesion segmentation with2D RECIST slices to learn slice-level segmentation information; 3) furtherfine-tune the Swin-transformer encoder to perform 3D lesion segmentation withlabeled 3D volumes. The network's performance is evaluated by the Dicesimilarity coefficient (DSC) and Hausdorff distance (HD) using an internal 3Dlesion dataset with 593 lesions extracted from multiple anatomical locations.The proposed MDU-ST demonstrates significant improvement over the competingmodels. The proposed method can be used to conduct automated 3D lesionsegmentation to assist radiomics and tumor growth modeling studies. This paperhas been accepted by the IEEE International Symposium on Biomedical Imaging(ISBI) 2023.

Multi-dimension unified Swin Transformer for 3D Lesion Segmentation in Multiple Anatomical Locations