Enabling longitudinal assessment of ankle-foot orthosis efficacy for children with cerebral palsy

Abstract

Ankle-foot orthoses (AFOs) are often prescribed to individuals with walking disabilities, including children with cerebral palsy. Despite the widespread use of AFOs, their efficacy is not well evaluated because the quantitative assessment of gait improvement from AFOs is currently limited to short-term, in-clinic observation. To better understand how AFOs perform in aiding individuals with walking disabilities and to further enhance their efficacy, longitudinal, continuous, non-invasive measurement is necessary. Ankle joint angle is a key parameter impacted by the AFO and is central to assessing AFO efficacy. With a wireless inertial body sensor network (BSN) mounted on -- or even embedded in -- the AFOs, the ankle joint angle can be extracted and then used to derive other gait parameters such as the range of ankle angle and the percentage of time in dorsi-/plantar-flexion mode. The methodology of extracting ankle joint angle and related gait parameters for assessing AFO efficacy is detailed in this paper. In order to obtain accurate spatial information, techniques for compensating integration drift, mounting error and multi-plane motion are also presented. The BSN results are validated against the industrial standard optical motion capture system on four children with cerebral palsy. An ankle angle RMSE of 2.41 degrees was achieved, demonstrating the potential of using BSNs for longitudinal assessment of AFO efficacy.