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/

Zendy Plus

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

Zendy Tools

Zendy Open

We describe a new multiresolution "nested encoder-decoder" convolutionalnetwork architecture and use it to annotate morphological patterns inreflectance confocal microscopy (RCM) images of human skin for aiding cancerdiagnosis. Skin cancers are the most common types of cancers, melanoma beingthe deadliest among them. RCM is an effective, non-invasive pre-screening toolfor skin cancer diagnosis, with the required cellular resolution. However,images are complex, low-contrast, and highly variable, so that cliniciansrequire months to years of expert-level training to be able to make accurateassessments. In this paper, we address classifying 4 key clinically importantstructural/textural patterns in RCM images. The occurrence and morphology ofthese patterns are used by clinicians for diagnosis of melanomas. The largesize of RCM images, the large variance of pattern size, the large-scale rangeover which patterns appear, the class imbalance in collected images, and thelack of fully-labeled images all make this a challenging problem to address,even with automated machine learning tools. We designed a novel nested U-netarchitecture to cope with these challenges, and a selective loss function tohandle partial labeling. Trained and tested on 56 melanoma-suspicious,partially labeled, 12k x 12k pixel images, our network automatically annotateddiagnostic patterns with high sensitivity and specificity, providing consistentlabels for unlabeled sections of the test images. Providing such annotationwill aid clinicians in achieving diagnostic accuracy, and perhaps moreimportant, dramatically facilitate clinical training, thus enabling much morerapid adoption of RCM into widespread clinical use process. In addition, ouradaptation of U-net architecture provides an intrinsically multiresolution deepnetwork that may be useful in other challenging biomedical image analysisapplications.

A Multiresolution Convolutional Neural Network with Partial Label Training for Annotating Reflectance Confocal Microscopy Images of Skin