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

This study proposes an unsupervised sequence-to-sequence learning approachthat automatically assesses the motion-induced reliability degradation of thecardiac volume signal (CVS) in multi-channel electrical impedance-basedhemodynamic monitoring. The proposed method attempts to tackle shortcomings inexisting learning-based assessment approaches, such as the requirement ofmanual annotation for motion influence and the lack of explicit mechanisms forrealizing motion-induced abnormalities under contextual variations in CVS overtime. By utilizing long-short term memory and variational auto-encoderstructures, an encoder--decoder model is trained not only to self-reproduce aninput sequence of the CVS but also to extrapolate the future in a parallelfashion. By doing so, the model can capture contextual knowledge lying in atemporal CVS sequence while being regularized to explore a general relationshipover the entire time-series. A motion-influenced CVS of low-quality isdetected, based on the residual between the input sequence and its neuralrepresentation with a cut--off value determined from the two-sigma rule ofthumb over the training set. Our experimental observations validated twoclaims: (i) in the learning environment of label-absence, assessmentperformance is achievable at a competitive level to the supervised setting, and(ii) the contextual information across a time series of CVS is advantageous foreffectively realizing motion-induced unrealistic distortions in signalamplitude and morphology. We also investigated the capability as apseudo-labeling tool to minimize human-craft annotation by preemptivelyproviding strong candidates for motion-induced anomalies. Empirical evidencehas shown that machine-guided annotation can reduce inevitable human-errorsduring manual assessment while minimizing cumbersome and time-consumingprocesses.

Unsupervised sequence-to-sequence learning for automatic signal quality assessment in multi-channel electrical impedance-based hemodynamic monitoring