PrimSeq: A deep learning-based pipeline to quantitate rehabilitation training
Open Access
- 16 June 2022
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS Digital Health
- Vol. 1 (6) , e0000044
Abstract
Stroke rehabilitation seeks to accelerate motor recovery by training functional activities, but may have minimal impact because of insufficient training doses. In animals, training hundreds of functional motions in the first weeks after stroke can substantially boost upper extremity recovery. The optimal quantity of functional motions to boost recovery in humans is currently unknown, however, because no practical tools exist to measure them during rehabilitation training. Here, we present PrimSeq, a pipeline to classify and count functional motions trained in stroke rehabilitation. Our approach integrates wearable sensors to capture upper-body motion, a deep learning model to predict motion sequences, and an algorithm to tally motions. The trained model accurately decomposes rehabilitation activities into elemental functional motions, outperforming competitive machine learning methods. PrimSeq furthermore quantifies these motions at a fraction of the time and labor costs of human experts. We demonstrate the capabilities of PrimSeq in previously unseen stroke patients with a range of upper extremity motor impairment. We expect that our methodological advances will support the rigorous measurement required for quantitative dosing trials in stroke rehabilitation.Keywords
All Related Versions
Funding Information
- American Heart Association (19AMTG35210398)
- U.S. National Library of Medicine (R01LM013316)
- U.S. National Library of Medicine (R01LM013316)
- National Institute of Neurological Disorders and Stroke (K02NS104207)
- National Center for Advancing Translational Sciences (UL1TR001445)
- Division of Graduate Education (NRT-HDR 1922658)
This publication has 47 references indexed in Scilit:
- Recovery after brain injury: mechanisms and principlesFrontiers in Human Neuroscience, 2013
- A role for ephrin-A5 in axonal sprouting, recovery, and activity-dependent plasticity after strokeProceedings of the National Academy of Sciences, 2012
- Activity recognition using cell phone accelerometersACM SIGKDD Explorations Newsletter, 2011
- Translating Animal Doses of Task-Specific Training to People With Chronic Stroke in 1-Hour Therapy Sessions: A Proof-of-Concept StudyNeurorehabilitation and Neural Repair, 2010
- Observation of Amounts of Movement Practice Provided During Stroke RehabilitationArchives of Physical Medicine and Rehabilitation, 2009
- Very Early Constraint-Induced Movement during Stroke Rehabilitation (VECTORS)Neurology, 2009
- Effects of Motor Training on the Recovery of Manual Dexterity After Primary Motor Cortex Lesion in Macaque MonkeysJournal of Neurophysiology, 2008
- Quaternion-Based Extended Kalman Filter for Determining Orientation by Inertial and Magnetic SensingIEEE Transactions on Biomedical Engineering, 2006
- Motor control of flexible octopus armsNature, 2005
- On the use of the I0-sinh window for spectrum analysisIEEE Transactions on Acoustics, Speech, and Signal Processing, 1980