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

Intervertebral disc disease, a prevalent ailment, frequently leads tointermittent or persistent low back pain, and diagnosing and assessing of thisdisease rely on accurate measurement of vertebral bone and intervertebral discgeometries from lumbar MR images. Deep neural network (DNN) models may assistclinicians with more efficient image segmentation of individual instances(disks and vertebrae) of the lumbar spine in an automated way, which is termedas instance image segmentation. In this work, we proposed SymTC, an innovativelumbar spine MR image segmentation model that combines the strengths ofTransformer and Convolutional Neural Network (CNN). Specifically, we designed aparallel dual-path architecture to merge CNN layers and Transformer layers, andwe integrated a novel position embedding into the self-attention module ofTransformer, enhancing the utilization of positional information for moreaccurate segmentation. To further improves model performance, we introduced anew data augmentation technique to create synthetic yet realistic MR imagedataset, named SSMSpine, which is made publicly available. We evaluated ourSymTC and the other 15 existing image segmentation models on our privatein-house dataset and the public SSMSpine dataset, using two metrics, DiceSimilarity Coefficient and 95% Hausdorff Distance. The results show that ourSymTC has the best performance for segmenting vertebral bones andintervertebral discs in lumbar spine MR images. The SymTC code and SSMSpinedataset are available at https://github.com/jiasongchen/SymTC.

SymTC: A Symbiotic Transformer-CNN Net for Instance Segmentation of Lumbar Spine MRI