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

Respiratory diseases kill million of people each year. Diagnosis of thesepathologies is a manual, time-consuming process that has inter andintra-observer variability, delaying diagnosis and treatment. The recentCOVID-19 pandemic has demonstrated the need of developing systems to automatizethe diagnosis of pneumonia, whilst Convolutional Neural Network (CNNs) haveproved to be an excellent option for the automatic classification of medicalimages. However, given the need of providing a confidence classification inthis context it is crucial to quantify the reliability of the model'spredictions. In this work, we propose a multi-level ensemble classificationsystem based on a Bayesian Deep Learning approach in order to maximizeperformance while quantifying the uncertainty of each classification decision.This tool combines the information extracted from different architectures byweighting their results according to the uncertainty of their predictions.Performance of the Bayesian network is evaluated in a real scenario wheresimultaneously differentiating between four different pathologies: control vsbacterial pneumonia vs viral pneumonia vs COVID-19 pneumonia. A three-leveldecision tree is employed to divide the 4-class classification into threebinary classifications, yielding an accuracy of 98.06% and overcoming theresults obtained by recent literature. The reduced preprocessing needed forobtaining this high performance, in addition to the information provided aboutthe reliability of the predictions evidence the applicability of the system tobe used as an aid for clinicians.

Uncertainty-driven ensembles of deep architectures for multiclass classification. Application to COVID-19 diagnosis in chest X-ray images