Pediatric Anklebot Hermano I. Krebs Senior Member, IEEE, Stefano Rossi, Seung-Jae Kim, Panagiotis K. Artemiadis, Dustin Williams, Enrico Castelli and Paolo Cappa Abstract—In this paper we present the alpha-prototype of a novel pediatric ankle robot. This lower-extremity robotic therapy module was developed at MIT to aid recovery of ankle function in children with cerebral palsy ages 5 to 8 years old. This lower-extremity robotic module will commence pilot testing with children with cerebral palsy at Blythedale Childrens Hospital (Valhalla, NY), Bambino Gesu Children’s Hospital (Rome, Italy), Riley Children’s Hospital (Indianapolis, IN). Its design follows the same guidelines as our upper-extremity robots and adult anklebot designs, i.e. it is a low friction, backdriveable device with intrinsically low mechanical impedance. We show the ankle robot characteristics and stability range. We also present pilot data with healthy children to demonstrate the potential of this device. I. I NTRODUCTION Cerebral palsy (CP) affects at least 2 in 1,000 children born in the United States. Studies have shown that in the US at least 5,000 infants and toddlers and 1,200 - 1,500 preschoolers are diagnosed with CP each year as developmental and mo- tor delays become more apparent (http://www.about-cerebral- palsy.org). These numbers are expected to grow worldwide with the increased survival of pre-term babies. Spastic diplegia and hemiplegic CP are the most common syndromes in children born at term and preterm infants [1]. CP habilitation is a process that seeks to enable the child to fully participate in society, with physical and occupational therapy playing a major role. The motivation behind such therapy is best ex- pressed by Hebbian ideas of nervous system plasticity, mainly This work is supported by the Cerebral Palsy International Research Foundation (CPIRF) and the Niarchos Foundation. Prof Cappa acknowledges the financial support of the Italian Health Ministry (Grant 2009 “Pilot study on a novel typology of medical devices: exoskeleton for pediatric rehabilitation).” Dr. H. I. Krebs is a co-inventor in the MIT-held patent for the robotic device used in this work. He holds equity positions in Interactive Motion Technologies, the company that manufactures this type of technology under license to MIT. H. I. Krebs is with the Department of Mechanical Engineering, Mas- sachusetts Institute of Technology, Cambridge, MA, USA, with the Depart- ment of Neurology and Neuroscience, Weill Medical College, Cornell Uni- versity, Burke Medical Research Institute, White Plains, NY, USA, and with the Department of Neurology, University of Maryland, School of Medicine, Baltimore, MD, USA (email H. I. Krebs: hikrebs@mit.edu). Stefano Rossi is with “Sapienza” University of Rome, Department of Mechanical and Aerospace Engineering, Rome, Italy. Seung-Jae Kim and Panagiotis K. Artemiadis are Postdoctoral Research Associates in the Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA. Dustin Williams is the Vice-President of Interactive Motion Technologies, Watertown, MA, USA. Enrico Castelli is with Bambino Gesu Children’s Hospital, Department of Neuroscience and Neurorehabilitation, Rome, Italy. Paolo Cappa is with “Sapienza” University of Rome, Department of Mechanical and Aerospace Engineering, Rome, Italy. that neurons that “fire” together, “wire” together. The human brain is capable of self-organization, or neuroplasticity [2, 3], so that learning offers an opportunity for motor recovery [4, 5]. Generally, therapy involves physical interaction with one or more therapists who assist and encourage the child through a number of repetitive exercises. The repetitive nature of therapy makes it amenable to being administered by properly designed robots. A robotic therapist can eliminate unnecessary exertion by the therapist, deliver a highly reproducible motor learning experience, quantitatively monitor and adapt to the child’s progress, and ensure consistency in planning a therapy program. Of course, one must take previous statement with the appropriate caveats as we do not know yet what constitutes optimal therapy for a particular individual needs. A pioneer of its class, MIT-MANUS, a robotic upper-limb manipulandum for shoulder and elbow training, was completed in 1991 [6]. Clinical trials involving MIT-MANUS have shown that robot-aided neuro-rehabilitation has a positive impact in stroke rehabilitation, reducing impairment, improving function and quality of life in both stroke inpatients and outpatients without increasing total cost [5]-[16]. This has motivated the development of new modules designed for rehabilitation of anti-gravity movements of the wrist, of the hand, and of the ankle [15]. In this paper, we report on development of a novel pediatric anklebot for children ages 5 to 8. We decided to focus on the ankle for two main reasons: first, in unimpaired subjects the ankle is the largest source of mechanical power during terminal stance [16], as the plantarflexors stabilize the forefoot rocker action [17]; second, the increase of plantarflexors muscular tone, with secondary equinus foot, is one of the main cause of gait impairment in CP [18]. The plantarflexors contribute as much as 50% of positive mechanical work in a single stride to enable forward propulsion [19]-[22]. In pre-swing, plan- tarflexors also act to advance the leg into swing phase while promoting knee flexion at toe-off [17]. Impairment at the ankle joint is of particular importance in CP. In some youngsters, it manifests as “equinus foot”, which is a simple name for a complex problem. The foot needs to clear the ground during the swing phase of gait and it needs to have a controlled landing during heel strike. It manifests itself as Equinus gait (True or Apparent) that, if allowed to mature as the child matures, can only be corrected through invasive orthopedic surgery. At present, equinus foot is typically addressed in the clinic via an ankle foot orthosis (AFO) that restricts the ankle’s range of motion. Hence a pediatric anklebot for children with CP ages 5 to 8