Perceptions of stroke survivors regarding factors affecting adoption of technology and exergames for rehabilitation.

BACKGROUND: Task-specific motor training and repetitive practice are essential components of clinical rehabilitation. Emerging evidence suggests that incorporating gaming interfaces (also referred to as "exergames"), including virtual reality and augmented reality (VR/AR)-based interfaces for motor training, can enhance the engagement and efficacy of poststroke rehabilitation. OBJECTIVE: To investigate perceptions of individuals with stroke regarding technology and exergames for rehabilitation. DESIGN: This qualitative phenomenological study included a convenience sample of 11 individuals with stroke (61.7 ± 12.4 years, 6 women and 5 men, 63.5 ± 41.2 months post stroke). SETTING: Community. INTERVENTIONS: N/A. OUTCOME MEASURES: Semistructured open-ended focus-group interviews to understand their perceptions on technology and exergames to improve recovery were coded using thematic content analysis. RESULTS: Individuals with stroke were comfortable using smartphones, computers, and rehabilitation technologies but had limited experiences using exergames and VR/AR devices. Individuals with stroke were motivated to use technologies and exergames to improve their functional recovery. Participants identified facilitators (eg, enhancing functional recovery, feedback, therapist supervision) and barriers (eg, safety, inaccessibility, inadequate knowledge) to adopting exergames in their daily lives. Participants wanted the exergames to be customizable, goal oriented, and enjoyable to maintain their engagement. They were willing to use exergames to improve their functional recovery but indicated that these games could not replace the therapist's supervision. CONCLUSIONS: Despite having limited experiences with exergames, people post stroke perceived that exergames could promote functional recovery. The perspectives gained from the present study can inform user-centered game design for neurorehabilitation.

Effects of object size and task goals on reaching kinematics in a non-immersive virtual environment.

Object size (large vs. small) and task goal (reach at a comfortable pace vs. reach as fast as possible) are well-accepted task constraints that influence reaching kinematics. However, it is unclear whether these two constraints affect reaching in the virtual environment (VE). The purpose of this study was to examine the effects of these two task constraints in VE. Forty-one healthy adults (11 males; mean age 24.9 years old) participated in this study. All adults were asked to play bubble popping games that ran on our virtual reality (VR) platform (SuperPop VR™), which can precisely measure reaching kinematics (including duration, straightness, jerkiness, and speed) in real time. Participants were blinded to the study purposes, and two task constraints were manipulated: bubble size (small vs. large) and goal (comfortable vs. fast). Participants began with their dominant hand for each condition and alternated hands after each condition was tested three times. A repeated ANOVA was used for analyses. Reaches for small bubbles had longer duration, faster speed, and jerkier with more curved trajectories than reaches for large bubbles; reaches for small bubbles also had larger elbow and shoulder range of motion than reaches for large bubbles. In addition, reaches during participants' comfortable pace had longer duration, slower speed, and larger shoulder range of motion than reaches during their fast pace. Additionally, participants' non-dominant hands had faster speed than their dominant hands. Our findings confirmed that object size and task goals affected reaching kinematics even during VR games. When designing a VR evaluation and intervention program for clinical populations, it is important to consider the virtual object size and task goals as factors in influencing a participant's performance.

Automated assessment of infant motor development to predict infant age: The determination of objective metrics of spontaneous kicking.

Though early intervention can improve outcomes for children with motor disabilities, delays in diagnosis can impact the success of intervention programs. Prior work indicates that spontaneous kicking patterns can be used to model typical infant motor development to assist in the early detection of motor delays. However, abnormalities in spontaneous movements are not well defined or readily observable through traditional functional assessments. In this research, a method is introduced for the early detection of delays through the assessment of spontaneous kicking data gathered using a wearable sensing suit. We present formulations of kinematic features identified in the clinical space, identify which features are significant predictors of infant age, and establish normative values. Finally, we offer an analysis of preterm (PT) infant data compared to normative values derived from term infants. Term and PT infants ranging in age from 1 to 10 months were studied. We found that frequency, duration, acceleration, inter-joint coordination, and maximum joint excursion metrics had a significant correlation with age. From these features, models of typical kicking development were created using data from term, typically developing infants. When compared to normative trends, PT infants display differing developmental trends.

Motivating Spontaneous Infant Kicking Motions through Long Term Learning Utilizing a Robotic Mobile System.

Our research investigates methods and systems to allow for early detection of motor impairment in infants and innovative interventions with the goal of improving longterm outcomes. A robotic baby mobile is utilized to motivate spontaneous kicking motions, which is used as a marker for predicting the potential of motor development delays. Our previous work investigated how the different stimuli modalities of a baby mobile can encourage infant kicking. We utilized a 3D camera system to detect the kicking motions, as well as recorded specific metrics of each kicking episode. In this work, we investigate the possibility of an infant having a preference of baby mobile stimuli that results in increased and sustained kicking motions. This preference is learned over multiple sessions with one infant and utilizes a Markov Decision Process to develop a policy.

Child-Robot Interaction in a Musical Dance Game: An Exploratory Comparison Study between Typically Developing Children and Children with Autism.

Using robots in therapy for children on the autism spectrum is a promising avenue for child-robot interaction, and one that has garnered significant interest from the research community. After preliminary interviews with stakeholders and evaluating music selections, twelve typically developing (TD) children and three children with Autism Spectrum Disorder (ASD) participated in an experiment where they played the dance freeze game to four songs in partnership with either a NAO robot or a human partner. Overall, there were significant differences between TD children and children with ASD (e.g., mimicry, dance quality, & game play). There were mixed results for TD children, but they tended to show greater engagement with the researcher. However, objective results for children with ASD showed greater attention and engagement while dancing with the robot. There was little difference in game performance between partners or songs for either group. However, upbeat music did encourage greater movement than calm music. Using a robot in a musical dance game for children with ASD appears to show the advantages and potential just as in previous research efforts. Implications and future research are discussed with the results.

Emerging challenges in AI and the need for AI ethics education.

Artificial Intelligence (AI) is reshaping the world in profound ways; some of its impacts are certainly beneficial but widespread and lasting harms can result from the technology as well. The integration of AI into various aspects of human life is underway, and the complex ethical concerns emerging from the design, deployment, and use of the technology serves as a reminder that it is time to revisit what future developers and designers, along with professionals, are learning when it comes to AI. It is of paramount importance to train future members of the AI community, and other stakeholders as well, to reflect on the ways in which AI might impact people's lives and to embrace their responsibilities to enhance its benefits while mitigating its potential harms. This could occur in part through the fuller and more systematic inclusion of AI ethics into the curriculum. In this paper, we briefly describe different approaches to AI ethics and offer a set of recommendations related to AI ethics pedagogy.

Robots are not immune to bias and injustice.

Human-human social constructs drive human-robot interactions; robotics is thus intertwined with issues surrounding inequity and racial injustices.

Number of trials necessary to achieve performance stability in a reaching kinematics movement analysis game.

STUDY DESIGN: A cross-sectional design was used. INTRODUCTION: Motion analysis is commonly used to analyze reaching movements of individuals, but how many trials are necessary to provide stable data? PURPOSE OF THE STUDY: The purpose of this study was to determine the number of trials necessary to obtain stable kinematic variables during reaching movements in a virtual environment. METHODS: Thirty children (10.39 ± 1.68 years) and 20 adults (26.55 ± 3.12 years) participated. All participants interacted with a virtual reality evaluation game called Super Pop VR to collect their reaching kinematics. The players were instructed to pop several sets of virtual bubbles at a self-selected pace and their upper-body coordinates were recorded and reaching kinematics were computed. Two methods (sequential averaging techniques, SAT, and intraclass correlation coefficient, ICC) were carried out to assess performance stability of each kinematic variable. RESULTS: For the SAT method, children exhibited 3.11 to 5.96 trial range to achieve stability in the dominant hand and 2.7 to 5.96 trial range in the nondominant hand, whereas adults exhibited 2.65 to 6.16 trial range in the dominant hand and 3.40 to 6.05 range in the nondominant hand. For the ICC method, to reach the "excellent" value of ICC, children's dominant hand required 4.20 trials and their nondominant hand required 2.5 trials, whereas adults' dominant hand required 3.00 trials and their nondominant hand required 2.80 trials. DISCUSSION AND CONCLUSIONS: Our findings supported the importance of determining the minimum number of trials required to obtain stable movements regardless of how familiar the movement seems to participants or how sensitive the apparatus is.

A Robotic Framework to Facilitate Sensory Experiences for Children with Autism Spectrum Disorder: A Preliminary Study.

The diagnosis of Autism Spectrum Disorder (ASD) in children is commonly accompanied by a diagnosis of sensory processing disorders. Abnormalities are usually reported in multiple sensory processing domains, showing a higher prevalence of unusual responses, particularly to tactile, auditory and visual stimuli. This paper discusses a novel robot-based framework designed to target sensory difficulties faced by children with ASD in a controlled setting. The setup consists of a number of sensory stations, together with two different robotic agents that navigate the stations and interact with the stimuli. These stimuli are designed to resemble real world scenarios that form a common part of one's everyday experiences. Given the strong interest of children with ASD in technology in general and robots in particular, we attempt to utilize our robotic platform to demonstrate socially acceptable responses to the stimuli in an interactive, pedagogical setting that encourages the child's social, motor and vocal skills, while providing a diverse sensory experience. A preliminary user study was conducted to evaluate the efficacy of the proposed framework, with a total of 18 participants (5 with ASD and 13 typically developing) between the ages of 4 and 12 years. We derive a measure of social engagement, based on which we evaluate the effectiveness of the robots and sensory stations in order to identify key design features that can improve social engagement in children.

Detection of Infant Motor Activity During Spontaneous Kicking Movements for Term and Preterm Infants Using Inertial Sensors.

Spontaneous kicking in infants is one of the earliest displays of motor skills. Abnormalities observed in these displays are an important indicator of later abnormal neuromotor function. However, these abnormalities are not well defined and difficult to detect outside of direct clinical observation. To allow for extended, non-clinical observation of spontaneous kicking, IMU sensors are attached to the limb segments of the infant's legs. An activity detection algorithm is then used to quantify kicking activity derived from collected measurement data. This paper presents our method in detail and discusses results from kicking data acquired from term and low-risk preterm infants.

The Ugly Truth About Ourselves and Our Robot Creations: The Problem of Bias and Social Inequity.

Recently, there has been an upsurge of attention focused on bias and its impact on specialized artificial intelligence (AI) applications. Allegations of racism and sexism have permeated the conversation as stories surface about search engines delivering job postings for well-paying technical jobs to men and not women, or providing arrest mugshots when keywords such as "black teenagers" are entered. Learning algorithms are evolving; they are often created from parsing through large datasets of online information while having truth labels bestowed on them by crowd-sourced masses. These specialized AI algorithms have been liberated from the minds of researchers and startups, and released onto the public. Yet intelligent though they may be, these algorithms maintain some of the same biases that permeate society. They find patterns within datasets that reflect implicit biases and, in so doing, emphasize and reinforce these biases as global truth. This paper describes specific examples of how bias has infused itself into current AI and robotic systems, and how it may affect the future design of such systems. More specifically, we draw attention to how bias may affect the functioning of (1) a robot peacekeeper, (2) a self-driving car, and (3) a medical robot. We conclude with an overview of measures that could be taken to mitigate or halt bias from permeating robotic technology.

Effect of feedback from a socially interactive humanoid robot on reaching kinematics in children with and without cerebral palsy: A pilot study.

PURPOSE: To examine whether children with or without cerebral palsy (CP) would follow a humanoid robot's (i.e., Darwin) feedback to move their arm faster when playing virtual reality (VR) games. METHODS: Seven children with mild CP and 10 able-bodied children participated. Real-time reaching was evaluated by playing the Super Pop VRTM system, including 2-game baseline, 3-game acquisition, and another 2-game extinction. During acquisition, Darwin provided verbal feedback to direct the child to reach a kinematically defined target goal (i.e., 80% of average movement time in baseline). Outcome variables included the percentage of successful reaches ("% successful reaches"), movement time (MT), average speed, path, and number of movement units. RESULTS: All games during acquisition and extinction had larger "%successful reaches," faster speeds, and faster MTs than the 2 games during baseline (p < .05). CONCLUSION: Children with and without CP could follow the robot's feedback for changing their reaching kinematics when playing VR games.

The effects of auditory and visual cues on timing synchronicity for robotic rehabilitation.

In this paper, we explore how the integration of auditory and visual cues can help teach the timing of motor skills for the purpose of motor function rehabilitation. We conducted a study using Amazon's Mechanical Turk in which 106 participants played a virtual therapy game requiring wrist movements. To validate that our results would translate to trends that could also be observed during robotic rehabilitation sessions, we recreated this experiment with 11 participants using a robotic wrist rehabilitation system as means to control the therapy game. During interaction with the therapy game, users were asked to learn and reconstruct a tapping sequence as defined by musical notes flashing on the screen. Participants were divided into 2 test groups: (1) control: participants only received visual cues to prompt them on the timing sequence, and (2) experimental: participants received both visual and auditory cues to prompt them on the timing sequence. To evaluate performance, the timing and length of the sequence were measured. Performance was determined by calculating the number of trials needed before the participant was able to master the specific aspect of the timing task. In the virtual experiment, the group that received visual and auditory cues was able to master all aspects of the timing task faster than the visual cue only group with p-values < 0.05. This trend was also verified for participants using the robotic arm exoskeleton in the physical experiment.

DEVELOPMENT AND USABILITY TESTING OF A REMOTE CONTROL APP FOR AN INTERACTIVE ROBOT.

Experimenters need robots that are easier to control for experimental purposes. In this paper, we conducted interviews for eliciting interaction requirements for human-robot interaction scenarios. User input was then incorporated into an Android application for remotely controlling an Aldebaran Nao robot for use in Wizard-of-Oz experiments and demos. The app was used in a usability study to compare it with an existing Nao remote control app. Results were positive, highlighting the ease-of-use and organization of the app. Future work includes a more complete usability trial evaluating the unique functionality of the app, as well as a case study of the app in a real Wizard-of-Oz experiment.

The Grad Cohort Workshop: Evaluating an Intervention to Retain Women Graduate Students in Computing.

Women engaged in computing career tracks are vastly outnumbered by men and often must contend with negative stereotypes about their innate technical aptitude. Research suggests women's marginalized presence in computing may result in women psychologically disengaging, and ultimately dropping out, perpetuating women's underrepresentation in computing. To combat this vicious cycle, the Computing Research Association's Committee on the Status of Women in Computing Research (CRA-W) runs a multi-day mentorship workshop for women graduate students called Grad Cohort, which consists of a speaker series and networking opportunities. We studied the long-term impact of Grad Cohort on women Ph.D. students' (a) dedication to becoming well-known in one's field, and giving back to the community (professional goals), (b) the degree to which one feels computing is an important element of "who they are" (computing identity), and (c) beliefs that computing skills are innate (entity beliefs). Of note, entity beliefs are known to be demoralizing and can lead to disengagement from academic endeavors. We compared a propensity score matched sample of women and men Ph.D. students in computing programs who had never participated in Grad Cohort to a sample of past Grad Cohort participants. Grad Cohort participants reported interest in becoming well-known in their field to a greater degree than women non-participants, and to an equivalent degree as men. Also, Grad Cohort participants reported stronger interest in giving back to the community than their peers. Further, whereas women non-participants identified with computing to a lesser degree than men and held stronger entity beliefs than men, Grad Cohort participants' computing identity and entity beliefs were equivalent to men. Importantly, stronger entity beliefs predicted a weaker computing identity among students, with the exception of Grad Cohort participants. This latter finding suggests Grad Cohort may shield students' computing identity from the damaging nature of entity beliefs. Together, these findings suggest Grad Cohort may fortify women's commitment to pursuing computing research careers and move the needle toward greater gender diversity in computing.

Effects of robotic therapy on upper-extremity function in children with cerebral palsy: A systematic review.

OBJECTIVE: To systematically examine the effects of robotic therapy on upper extremity (UE) function in children with cerebral palsy (CP). METHODS: A systematic literature search was conducted in Pubmed, CINAHL, Cochrane, PsychInfo, TRIP, and Web of Science up to July 2013. Studies of children with CP, using robotic therapy and measures of UE were included. RESULTS: Nine articles using three different robotic systems were included. Of these, seven were case studies. Overall, robotic therapy showed the potential effects as all studies reported at least one positive outcome: a moderate effect in improving reaching duration, smoothness, or decreased muscle tone, and a small to large effect in standardized clinical assessment (e.g. Fugl-Meyer). CONCLUSION: This review confirms the potential for robotic therapy to improve UE function in children with CP. However, the paucity of group design studies summons the need for more rigorous research before conclusive recommendations can be made.

Lower limb pose estimation for monitoring the kicking patterns of infants.

Monitoring the spontaneous kicking patterns of infants can give insight into their development. A computer vision based method for estimating the pose of an infant's leg from range images is presented in this paper. After some manual inputs for initialization, the range data associated with the infant is extracted. The method uses Robust Point Set Registration (RPSR) to fit an articulated model to the subject in every frame in the sequence, thus it provides the joint trajectories over time of the kicking kinematics. For validation, the method is used to track the articulation of a robotic humanoid that was programmed to kick in a fashion similar to an infant. Furthermore, the method is applied to a sequence collected from an actual infant and the resultant signal estimates are presented.

Automated feet detection for clinical gait assessment.

The paper describes a computer vision method for estimating the clinical gait metrics of walking patients in unconstrained environments. The method employs background subtraction to produce a silhouette of the subject and a randomized decision forest to detect their feet. Given the feet detections, the stride and step length, cadence, and walking speed are estimated. Validation of the system is presented through an error analysis on manually annotated videos of subjects walking in different outdoor settings. This method is significant as it provides clinical therapists and non-specialists the opportunity to record from any camera and obtain high accuracy estimates of the clinical gait metrics for subjects walking at outdoor or at-home locations.

Effect of a Home-Based Virtual Reality Intervention for Children with Cerebral Palsy Using Super Pop VR Evaluation Metrics: A Feasibility Study.

Objective. The purpose of this pilot study was to determine whether Super Pop VR, a low-cost virtual reality (VR) system, was a feasible system for documenting improvement in children with cerebral palsy (CP) and whether a home-based VR intervention was effective. Methods. Three children with CP participated in this study and received an 8-week VR intervention (30 minutes × 5 sessions/week) using the commercial EyeToy Play VR system. Reaching kinematics measured by Super Pop VR and two fine motor tools (Bruininks-Oseretsky Test of Motor Proficiency second edition, BOT-2, and Pediatric Motor Activity Log, PMAL) were tested before, mid, and after intervention. Results. All children successfully completed the evaluations using the Super Pop VR system at home where 85% of the reaches collected were used to compute reaching kinematics, which is compatible with literature using expensive motion analysis systems. Only the child with hemiplegic CP and more impaired arm function improved the reaching kinematics and functional use of the affected hand after intervention. Conclusion. Super Pop VR proved to be a feasible evaluation tool in children with CP.