A new task-oriented assessment and training system for assessment of reach-to-target ability in stroke survivors: a feasibility study

We invented a new task-oriented assessment and training system based on opto-electronic systems for assessing and training upper extremity (UE) motor performance in people after stroke. Because kinematic data are a quantified measure of movements, the size of the workspace of the hand, the execution time, and the velocity of the movement as well as information about the movement's quality with respect to coordination, smoothness and other functional characteristics. Compared to the classic clinical scales, kinematic data are more objective, discriminative measure that quantifies movement biomechanics. In this trail, we just UTILIZE THIS NEW SYSTEM AS AN ASSESSMENT INSTRUMENT to obtain four kinematic variables of UE motor performance in order to investigate the psychometric properties of these kinematic outcome measures of two kinds of reaching performances. In other words, we just study discriminative and evaluative ability (the reproducibility and responsiveness) of these kinematic data regarding changes in UE movement control associated with motor recovery and/or responses to therapeutic interventions. We hope this study can contribute to a better understanding of the changes in the UE kinematic measures in relation to the changes in an individual’s clinical function by analyzing the results of treatments and assessing their efficacy, discriminating between healthy and pathology conditions in relation to detect functional impairments when the assessment methods for a new device is developed. In this trial, we DON’T USE THIS SYSTME AS AN INTERVENTION to alter or influence the clinical outcomes. We just observe a cohort of the sub acute phase after stroke and use this new system as an assessment instrument to examine and report on the changes in subjects’ UE function by four kinematic variables. Subjects in the study may receive therapeutic or other interventions, but we do not assign specific and controlled interventions to the subjects of this study. SYSTEM DESCRIPTION The system consists of a doctor station and a patient station. The doctor station is operated by the therapist in hospital and includes system software and a USB disk. Through the system software, therapists can perform the following tasks on computer: storage and query of patient information, evaluation of patient UE function, movement selection, and exercise parameter settings. The therapist stores the assessing and/or training plan into the USB disk for patients. Therapist can also review patient’s movements performance history saved in USB disk at the next clinical visit. This makes a loop that can cover the whole rehabilitation phases. Two kinds of functional tasks: nine self-assisted reaching movements and twenty-three active reaching movements make up the core of the movements of the system. But in this trial, two standardized reaching movements: the self-assisted reaching by shoulder flexion and the active reaching by shoulder flexion respectively represented two kinds of functional tasks are used for testing. The patient station has an infra-camera based main station and a set of accessories, including a TV as display device, a remote control and a set of retro-reflective markers. The markers will be put on patient's shoulder, elbow and hand for motion tracking purpose. During assessing and/or training session, the patient puts the infrared reflective markers on the affected UE, sitting on the direction of the main station at a certain distance. The USB disk is inserted into the main station which is connected with a TV to show the system interface. The main station has the following functions: 1. Manage patient’s kinematic data by loading the plan and saving the performance history in USB disk for performance review purpose. 2. Give audio/video instructions to teach patients how to use the system and how to perform movements. 3. Tracks patient’s upper limb movement in real time and displays upper limb and trunk movement in virtual scenarios. An infrared camera – retro reflective markers based video analysis module is used in patient training station for motion tracking. The video is captured in 25FPS (Frame Per Second) with resolution of 352*288 and for each frame, image processing and analysis algorithms are used to identify the shoulder/elbow/hand markers. After the markers are identified, a quantified measure of movements, the size of the workspace of the hand, the execution time, and the velocity of the movement as well as information about the movement's quality are calculated and recorded. All the data will be saved in USB disk for the next clinical review. PROTOCOL The study design involves a repeated-baseline assessment: a 3 days interval between the first (baseline 1) and second (baseline 2) baseline assessments as well as a 4 weeks poststroke assessment after the baseline 2. UE assessment includes a battery of clinical scales: the Fugl-Meyer Assessment (FMA) Scale, the Motor Status Score (MSS), the Action Research Arm Test(ARAT) and four kinematic outcome measures: movement time, maximum velocity, number of velocity peaks, hand path ratio. All subjects are divided into 3 subgroups: group A, group B and healthy controls based on the inclusion and exclusion criteria. Group A(40 subjects, affected UE with better function) will be assessed thrice: four kinematic variables for two reaching movements at baseline 1; four kinematic variables for two reaching movements and FMA, ARAT,MSS at baseline 2(3 days after baseline 1); four kinematic variables for two reaching movements and FMA, ARAT, MSS at the 4 weeks after baseline 2. Group B(40 subjects, affected UE with worse function) will be assessed twice: four kinematic variables for self-assisted reaching movement at baseline 1; four kinematic variables for self-assisted reaching movement and FMA, ARAT, MSS at baseline 2(3 days after baseline). Healthy controls (20 subjects, dominant arm) will be assessed twice: four kinematic variables for two reaching movements at baseline 1; four kinematic variables for two reaching movements at baseline 2. The objective of this study is to determine the reproducibility and responsiveness of UE kinematic metrics while performing two different reaching tasks in a prospective longitudinal cohort of the subacute phase after stroke. The relative reliability (to compute Interclass correlation coefficient, ICC) and the absolute reliability (to calculate the standard error of measure (SEM), the minimal detectable change (MDC)) within stable subjects (between baseline 1 and 2 for group A,group B and healthy controls) are used to evaluate reproducibility of UE kinematics. The minimally important change (MIC) values of UE kinematics in an observed longitudinal change (between baseline 2 and 4 weeks for group A, ARAT as an anchor) in UE motor performance and the link between clinical scales (FMA, ARAT, MSS) and four kinematic variables of motor performance (for group A and group B) are used to evaluate responsiveness of UE kinematics. KINEMATIC MOVEMENT ASSESSMENT During assessing session, the therapist puts the retro-reflective markers on the subjects’ shoulder, elbow and hand for motion tracking purpose. The subjects will sit in a height adjustable chair with their back against the chair back on the direction of the main station at the 1.8 meters. The chair will be adjusted for subjects while sitting to align shoulder marker and the infra-camera. The start position for the active reaching by shoulder flexion is with the tested UE resting on the ipsilateral side, such that the shoulder is in neutral position, the arm hanging straight down. After the calibration of the system, the subjects are instructed to flex the tested shoulder, maintaining the elbow in extension in order to reach forward toward the virtual target on the TV screen. Then return to starting position. The target is set at 80% of the subjects’ shoulder maximal range of motion. While the start position for the self-assisted reaching by shoulder flexion is with the tested hand holding by the healthy hand, such that the tested shoulder is in 45 degree flexion and 30 degree adduction, both hands resting between the two legs. After the calibration of the system, the subjects are instructed to flex the tested shoulder with the healthy hand assistance, maintaining both elbows in extension in order to reach forward toward the virtual target on the TV screen. Then return to starting position. The target is fixed at which the shoulder must to passively flex to 145 degree. All tasks will be performed without any constraint at a comfortable self-paced speed. Each task will be repeated five times. For each session, efforts will make to provide stability of test conditions across sessions, such as using identical instructions and data collection protocols and having the same trained physical therapist perform all assessments in the same location. Subjects are informed of the aim and given appropriate instructions before the test, in order to help them better understand the study. Subjects will be given the following instructions for the two kinds of reaching tasks: “Look at the target. When the system say ‘start,’ reach toward the target at your own pace. When you have reached as close to the target as you can, hold your position until the system ask you to return to the start position.” For each reaching task, subjects are provided 1 or 2 practice trials prior to recording to familiarize themselves with the tasks and the instructions. Each assessment session will take approximately 20 minutes. Motion tracking data are collected by an infra-camera and retro-reflective marker based motion tracking system. The system captures video in 20FPS (Frame Per Second) with resolution of 352*288 and for each frame, image processing and analysis algorithms are exploited to identify the positions of the shoulder, elbow and hand markers. The position data recorded is transferred to Matlab software for customized analysis.