The Role Of Wearable Wrist Inertial Sensors To Quantify Arm Asymmetry During Gait In Unilateral Spastic Cerebral Palsy (USCP)
Aviva Wolff, EdD, OTR, CHT1; Mark Lenhoff, BS1; Aaron Daluiski, MD2
1Hospital for Special Surgery, New York, NY, 2Hand and Upper Extremity Surgery, Hospital for Special Surgery, New York, NY
Individuals with USCP often seek intervention to improve arm motion and position during gait to address the problem of increased upper extremity spasticity that interferes with reciprocal arm swing, particularly during fast gait and running. Current quantitative measures that assess arm movements during gait are costly, cumbersome, and time-consuming. There is a need for more-efficient methods that are easily implementable in clinical settings. The purpose of this proof-of-concept study was to quantify arm movement and asymmetry during 3 walking speeds in individuals with USCP using light-weight wearable inertial sensors.
Materials and Methods:
9 individuals (6 male, 3 female) with USCP (age range: 3-29, mean age=12±8) wore wearable MTw inertial sensors on each wrist (XSens, Enschede, Netherlands) during 3 gait speeds: self-selected comfortable walking, self-selected fast walking, and running. Integrated acceleration data from 3 planes of arm movement were collected continuously at a sample rate of 100Hz during ambulation. The magnitude of the acceleration due to the force of gravity was subtracted from each trial. The maximum average acceleration was calculated for each arm over one second for each of the 3 conditions, and an asymmetry index (ASI) calculated.
Arm asymmetry increased across speed conditions (F2,23=15, p=0.003). Average ASI (0=perfect symmetry) increased to 43±17 during fast walking, and to 65±33 during running compared to 9±9 at baseline walking.
Our results that arm asymmetry increased in running and fast walking support the proof-of-concept that inertial sensors are able to quantify differences between arm movements and capture changes in asymmetry across gait speeds in individuals with USCP. These findings have broad implications for use in activity tracking and quantifying arm movements in a wide range of real-world activity in this population.
Back to 2019 Abstracts