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

Musculoskeletal and neurological disorders are the most common causes ofwalking problems among older people, and they often lead to diminished qualityof life. Analyzing walking motion data manually requires trained professionalsand the evaluations may not always be objective. To facilitate early diagnosis,recent deep learning-based methods have shown promising results for automatedanalysis, which can discover patterns that have not been found in traditionalmachine learning methods. We observe that existing work mostly applies deeplearning on individual joint features such as the time series of jointpositions. Due to the challenge of discovering inter-joint features such as thedistance between feet (i.e. the stride width) from generally smaller-scalemedical datasets, these methods usually perform sub-optimally. As a result, wepropose a solution that explicitly takes both individual joint features andinter-joint features as input, relieving the system from the need ofdiscovering more complicated features from small data. Due to the distinctivenature of the two types of features, we introduce a two-stream framework, withone stream learning from the time series of joint position and the other fromthe time series of relative joint displacement. We further develop a mid-layerfusion module to combine the discovered patterns in these two streams fordiagnosis, which results in a complementary representation of the data forbetter prediction performance. We validate our system with a benchmark datasetof 3D skeleton motion that involves 45 patients with musculoskeletal andneurological disorders, and achieve a prediction accuracy of 95.56%,outperforming state-of-the-art methods.

A Two-stream Convolutional Network for Musculoskeletal and Neurological Disorders Prediction