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

Lung cancer is the leading cause of cancer death worldwide and a goodprognosis depends on early diagnosis. Unfortunately, screening programs for theearly diagnosis of lung cancer are uncommon. This is in-part due to the at-riskgroups being located in rural areas far from medical facilities. Reaching thesepopulations would require a scaled approach that combines mobility, low cost,speed, accuracy, and privacy. We can resolve these issues by combining thechest X-ray imaging mode with a federated deep-learning approach, provided thatthe federated model is trained on homogenous data to ensure that no single datasource can adversely bias the model at any point in time. In this study we showthat an image pre-processing pipeline that homogenizes and debiases chest X-rayimages can improve both internal classification and external generalization,paving the way for a low-cost and accessible deep learning-based clinicalsystem for lung cancer screening. An evolutionary pruning mechanism is used totrain a nodule detection deep learning model on the most informative imagesfrom a publicly available lung nodule X-ray dataset. Histogram equalization isused to remove systematic differences in image brightness and contrast. Modeltraining is performed using all combinations of lung field segmentation, closecropping, and rib suppression operators. We show that this pre-processingpipeline results in deep learning models that successfully generalize anindependent lung nodule dataset using ablation studies to assess thecontribution of each operator in this pipeline. In stripping chest X-ray imagesof known confounding variables by lung field segmentation, along withsuppression of signal noise from the bone structure we can train a highlyaccurate deep learning lung nodule detection algorithm with outstandinggeneralization accuracy of 89% to nodule samples in unseen data.

Debiasing pipeline improves deep learning model generalization for X-ray based lung nodule detection