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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Importance: Machine learning (ML) approaches to facial landmark localizationcarry great clinical potential for quantitative assessment of facial functionas they enable high-throughput automated quantification of relevant facialmetrics from photographs. However, translation from research settings toclinical applications requires important improvements. Objective: To develop anML algorithm for accurate facial landmarks localization in photographs offacial palsy patients, and use it as part of an automated computer-aideddiagnosis system. Design, Setting, and Participants: Facial landmarks weremanually localized in portrait photographs of eight expressions obtained from200 facial palsy patients and 10 controls. A novel ML model for automatedfacial landmark localization was trained using this disease-specific database.Model output was compared to manual annotations and the output of a modeltrained using a larger database consisting only of healthy subjects. Modelaccuracy was evaluated by the normalized root mean square error (NRMSE) betweenalgorithms' prediction and manual annotations. Results: Publicly availablealgorithms provide poor results when applied to patients compared to healthycontrols (NRMSE, 8.56 +/- 2.16 vs. 7.09 +/- 2.34, p << 0.01). We foundsignificant improvement in facial landmark localization accuracy for theclinical population when using a model trained with a relatively small numberpatients' photographs (1440) compared to a model trained using several thousandmore images of healthy faces (NRMSE, 6.03 +/- 2.43 vs. 8.56 +/- 2.16, p <<0.01). Conclusions: Retraining a landmark detection model with a small numberof clinical images significantly improved landmark detection performance infrontal view photographs of the clinical population. These results representthe first steps towards an automatic system for computer-aided assessment infacial palsy.

Toward an Automatic System for Computer-Aided Assessment in Facial Palsy