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

Deep learning (DL) has drawn tremendous attention in object localization andrecognition for both natural and medical images. U-Net segmentation models havedemonstrated superior performance compared to conventional handcraftedfeature-based methods. Medical image modality-specific DL models are better attransferring domain knowledge to a relevant target task than those that arepretrained on stock photography images. This helps improve model adaptation,generalization, and class-specific region of interest (ROI) localization. Inthis study, we train chest X-ray (CXR) modality-specific U-Nets and otherstate-of-the-art U-Net models for semantic segmentation of tuberculosis(TB)-consistent findings. Automated segmentation of such manifestations couldhelp radiologists reduce errors and supplement decision-making while improvingpatient care and productivity. Our approach uses the publicly available TBX11KCXR dataset with weak TB annotations, typically provided as bounding boxes, totrain a set of U-Net models. Next, we improve the results by augmenting thetraining data with weak localizations, post-processed into an ROI mask, from aDL classifier that is trained to classify CXRs as showing normal lungs orsuspected TB manifestations. Test data are individually derived from the TBX11KCXR training distribution and other cross-institutional collections includingthe Shenzhen TB and Montgomery TB CXR datasets. We observe that our augmentedtraining strategy helped the CXR modality-specific U-Net models achievesuperior performance with test data derived from the TBX11K CXR trainingdistribution as well as from cross-institutional collections (p < 0.05).

Improved Semantic Segmentation of Tuberculosis-consistent findings in  Chest X-rays Using Augmented Training of Modality-specific U-Net Models with  Weak Localizations