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

Neural networks often obtain sub-optimal representations during training,which degrade robustness as well as classification performances. This is asevere problem in applying deep learning to bio-medical domains, since modelsare vulnerable to being harmed by irregularities and scarcities in data. Inthis study, we propose a pre-training technique that handles this challenge insleep staging tasks. Inspired by conventional methods that experiencedphysicians have used to classify sleep states from the existence ofcharacteristic waveform shapes, or template patterns, our method introduces acosine similarity based convolutional neural network to extract representativewaveforms from training data. Afterwards, these features guide a model toconstruct representations based on template patterns. Through extensiveexperiments, we demonstrated that guiding a neural network with templatepatterns is an effective approach for sleep staging, since (1) classificationperformances are significantly enhanced and (2) robustness in several aspectsare improved. Last but not least, interpretations on models showed that notablefeatures exploited by trained experts are correctly addressed during predictionin the proposed method.

TRIER: Template-Guided Neural Networks for Robust and Interpretable Sleep Stage Identification from EEG Recordings