Evaluation of computer vision based objective measures for complementary balance function description and assessment in multiple sclerosis
Título de la revista
ISSN de la revista
Título del volumen
Multiple Sclerosis is a neurodegenerative– autoimmune disorder caused by a demyelination process of the axonal tracts within the Central Nervous System. This condition will increasingly affect cognitive, perceptual, motor and even vital life functions at different rates. Sensorimotor impairments have an increasing impact in the patient functionality, altering basic abilities e.g. static and dynamic equilibrium, whose preservation is an important therapeutic goal. The assessment of the state and progression of the associated disabilities is a relevant issue in the election and adjustment of a rehabilitation pathway. This work presents an exploratory study on the use of angular kinematic variables as objective descriptors for Multiple Sclerosis diagnostic support, comparing its behavior against the score values for a subset of five equilibrium tests within the Berg Balance Scale. These values were estimated using a computer vision-based framework integrating data from a Kinect sensor and the NiTE skeleton model. This version of the framework provides angular measures for mediolateral and anteroposterior balance. To evaluate this quantitative approach, a sample of six patients with diagnosis of Multiple Sclerosis, and able to maintain the standing position, performed all five balance tests while both mediolateral and anteroposterior angles were registered along each of them and compare its behaviour against the corresponding evaluation using the standard Berg Balance Scale scores assigned by a physiotherapist. © Springer Nature Singapore Pte Ltd. 2017.
Biomedical engineering , Diagnosis , Function evaluation , Neurodegenerative diseases , Patient rehabilitation , Autoimmune disorder , Berg balance scale , Central nervous systems , Dynamic equilibria , Exploratory studies , Kinematic variables , Multiple sclerosis , Quantitative approach , Computer vision