Robotics and Autonomous Systems 91 (2017) 169–178 Contents lists available at ScienceDirect Robotics and Autonomous Systems journal homepage: www.elsevier.com/locate/robot Robotic wrist training after stroke: Adaptive modulation of assistance in pediatric rehabilitation Francesca Marini a , Charmayne M.L. Hughes b , Valentina Squeri a , Luca Doglio c , Paolo Moretti c , Pietro Morasso a , Lorenzo Masia d, * a Department of Robotics, Brain and Cognitive Sciences, Istituto Italiano di Tecnologia, Genova, Italy b Department of Kinesiology, San Francisco State University, 1600 Holloway Avenue, San Francisco, CA 94132, United States c Physical Medicine and Rehabilitation Institute G. Gaslini, Genova, Italy d Robotics Research Centre, School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Ave, Singapore 639798, Singapore highlights • Pediatric stroke leads to limb hemiparesis, sensory impairments, and spasticity. • A 14-year old stroke patient completed in a 3-month wrist robotic training program. • The robot provided online adaptive modulation of assistance instantaneously during each trial. • Robot therapy led to positive changes in upper limb motor coordination and function. • In addition, the patient needed less robot assistance to complete each trial. article info Article history: Available online 25 January 2017 Keywords: Motor dysfunction Robotic rehabilitation Pediatric stroke Wrist rehabilitation abstract In this paper we present a case study in which a 14-year-old, right-handed stroke patient with severe weakness, spasticity, and motor dysfunction of the left upper extremity participated in a three-month distal robotic training program. The robotic device was compliant to the patient’s movements and was able to modulate the level of assistance continuously throughout the trial (i.e., online adaptive modulation). Standard clinical and robotic evaluations of upper extremity motor performance were conducted before and after robotic training. There were improvements in upper extremity spasticity and motor functions. In addition, robotic training lead to positive changes in wrist active range of motion and kinematics: movements were smoother and there was a noticeable decrease in the level of robotic intervention required to complete each trial. In sum, results of the present case study demonstrate that distal upper extremity robotic rehabilitation that features the proposed adaptive control algorithm promoted positive changes in upper limb motor coordination and function after pediatric stroke. © 2017 Published by Elsevier B.V. 1. Introduction Stroke or cerebral vascular accident (CVA), caused by the oc- clusion or rupture of cerebral blood vessels, is the leading cause of neurological disability worldwide. It has been estimated that among the children who survive, between 50% and 80% will exhibit permanent upper extremity sensorimotor deficits [1]. The most common long-term impairments are hemiparesis (weakness of the entire left or right side of the body) or hemiplegia (total or partial paralysis on one side of the body) [2], which is often accompanied * Corresponding author. E-mail address: lorenzo.masia@ntu.edu.sg (L. Masia). by muscle weakness, impaired motor coordination, and impair- ments to sensory mechanisms of the affected paretic arm [3,4]. Functional loss of mobility due to hemiplegia has a significantly negative impact on the ability to perform activities of daily living, such as reaching and grasping and object manipulation. In addition to hemiplegia, stroke patients often present with spasticity and contractures of the elbow, wrist, and finger muscles [5,6]. Upper extremity spasticity can lead to pain, tendon retraction or muscle weakness, which has negative effects on a patient’s quality of life and rehabilitation outcomes [7,8]. Current conventional stroke rehabilitation therapies are a labor intensive process that involve daily one-on-one interactions with therapists that can last for several weeks. Fortunately, robotic advancements in the last decade have provided researchers http://dx.doi.org/10.1016/j.robot.2017.01.006 0921-8890/© 2017 Published by Elsevier B.V.