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

Statistical shape models (SSM) have been well-established as an excellenttool for identifying variations in the morphology of anatomy across theunderlying population. Shape models use consistent shape representation acrossall the samples in a given cohort, which helps to compare shapes and identifythe variations that can detect pathologies and help in formulating treatmentplans. In medical imaging, computing these shape representations from CT/MRIscans requires time-intensive preprocessing operations, including but notlimited to anatomy segmentation annotations, registration, and texturedenoising. Deep learning models have demonstrated exceptional capabilities inlearning shape representations directly from volumetric images, giving rise tohighly effective and efficient Image-to-SSM networks. Nevertheless, thesemodels are data-hungry and due to the limited availability of medical data,deep learning models tend to overfit. Offline data augmentation techniques,that use kernel density estimation based (KDE) methods for generatingshape-augmented samples, have successfully aided Image-to-SSM networks inachieving comparable accuracy to traditional SSM methods. However, theseaugmentation methods focus on shape augmentation, whereas deep learning modelsexhibit image-based texture bias resulting in sub-optimal models. This paperintroduces a novel strategy for on-the-fly data augmentation for theImage-to-SSM framework by leveraging data-dependent noise generation or textureaugmentation. The proposed framework is trained as an adversary to theImage-to-SSM network, augmenting diverse and challenging noisy samples. Ourapproach achieves improved accuracy by encouraging the model to focus on theunderlying geometry rather than relying solely on pixel values.

ADASSM: Adversarial Data Augmentation in Statistical Shape Models From Images