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

Recovery after stroke is often incomplete, but rehabilitation training maypotentiate recovery by engaging endogenous neuroplasticity. In preclinicalmodels of stroke, high doses of rehabilitation training are required to restorefunctional movement to the affected limbs of animals. In humans, however, thenecessary dose of training to potentiate recovery is not known. This ignorancestems from the lack of objective, pragmatic approaches for measuring trainingdoses in rehabilitation activities. Here, to develop a measurement approach, wetook the critical first step of automatically identifying functionalprimitives, the basic building block of activities. Forty-eight individualswith chronic stroke performed a variety of rehabilitation activities whilewearing inertial measurement units (IMUs) to capture upper body motion.Primitives were identified by human labelers, who labeled and segmented theassociated IMU data. We performed automatic classification of these primitivesusing machine learning. We designed a convolutional neural network model thatoutperformed existing methods. The model includes an initial module to computeseparate embeddings of different physical quantities in the sensor data. Inaddition, it replaces batch normalization (which performs normalization basedon statistics computed from the training data) with instance normalization(which uses statistics computed from the test data). This increases robustnessto possible distributional shifts when applying the method to new patients.With this approach, we attained an average classification accuracy of 70%.Thus, using a combination of IMU-based motion capture and deep learning, wewere able to identify primitives automatically. This approach builds towardsobjectively-measured rehabilitation training, enabling the identification andcounting of functional primitives that accrues to a training dose.

Towards data-driven stroke rehabilitation via wearable sensors and deep learning