AUTHOR=Campos Talita , Ratnadurai-Giridharan Shivakeshavan , Wairimu Marion-Victoria , Li Annie , Tamayo Sara , Friel Kathleen M. TITLE=Robotic bimanual arm training improves upper limb function in children with hemiparesis: a feasibility study JOURNAL=Frontiers in Pediatrics VOLUME=Volume 13 - 2025 YEAR=2025 URL=https://www.frontiersin.org/journals/pediatrics/articles/10.3389/fped.2025.1502481 DOI=10.3389/fped.2025.1502481 ISSN=2296-2360 ABSTRACT=IntroductionIn children with hemiparesis, bimanual training can produce greater gains in upper limb function compared with unimanual training of the impaired upper limb. Moreover, repetitive skilled practice is a critical component of effective interventions. The objective of this study was to determine whether upper limb training using a bimanual-to unimanual robotic training device is feasible and effective for improving hand function in children with hemiplegia. We hypothesized that this robotic training would improve motor outcomes on the Assisting Hand Assessment (AHA), the Box and Blocks Test (BBT), and the Jebsen-Taylor Test of Hand Function (JTTHF).MethodsChildren (n = 8, 6 males, 2 females, age 5-17 years) with hemiparesis participated in a feasibility study using a robotic device called the Bimanual Arm Trainer (BAT, Mirrored Motion Works). This device encourages mirrored bimanual movements and uses an engaging gaming interface to encourage repetitive movements. The BAT provides bimanual-to-unimanual training of yoked shoulder external rotation and elbow extension, as well as training of pronation, supination and grasp and release of each hand independently.ResultsAlthough the study was halted by the COVID-19 pandemic, eight children completed 18 training sessions on the BAT (2x/week for 9 weeks). While range of motion did not change significantly, children significantly improved in bimanual (AHA) function.DiscussionChildren enjoyed the device, and provided feedback that was used to improve the gaming environment. Further work is needed to determine ideal dosing to optimize improvements.