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

Freezing of gait (FoG) is one of the most common symptoms of Parkinson'sdisease, which is a neurodegenerative disorder of the central nervous systemimpacting millions of people around the world. To address the pressing need toimprove the quality of treatment for FoG, devising a computer-aided detectionand quantification tool for FoG has been increasingly important. As anon-invasive technique for collecting motion patterns, the footstep pressuresequences obtained from pressure sensitive gait mats provide a greatopportunity for evaluating FoG in the clinic and potentially in the homeenvironment. In this study, FoG detection is formulated as a sequentialmodelling task and a novel deep learning architecture, namely AdversarialSpatio-temporal Network (ASTN), is proposed to learn FoG patterns acrossmultiple levels. A novel adversarial training scheme is introduced with amulti-level subject discriminator to obtain subject-independent FoGrepresentations, which helps to reduce the over-fitting risk due to the highinter-subject variance. As a result, robust FoG detection can be achieved forunseen subjects. The proposed scheme also sheds light on improvingsubject-level clinical studies from other scenarios as it can be integratedwith many existing deep architectures. To the best of our knowledge, this isone of the first studies of footstep pressure-based FoG detection and theapproach of utilizing ASTN is the first deep neural network architecture inpursuit of subject-independent representations. Experimental results on 393trials collected from 21 subjects demonstrate encouraging performance of theproposed ASTN for FoG detection with an AUC 0.85.

Multi-level Adversarial Spatio-temporal Learning for Footstep Pressure based FoG Detection