Faculty Bibliography

Objective and accurate activity identification of physical activities in everyday life is an important aspect in assessing the impact of various post-stroke rehabilitation therapies and interventions. Since post-stroke hemiparesis affects gait and balance in individuals with stroke, activity identification algorithms that consider stroke-specific movement irregularities are needed. While wearable physical activity monitors provide the means to detect activities in the free-living, algorithms using their data are specific to the wear location of the device. This pilot study builds, validates, and compares three machine learning algorithms (linear support vector machine, Random Forest, and RUSBoosted trees) at three popular wear locations (wrist, waist, and ankle) to identify and accurately distinguish mobility-related activities (sitting, standing and walking) in individuals with chronic stroke. A total of 102 minutes of data from two lab visits of three-stroke participants was used to build the classifiers. A 5-fold cross-validation technique was used to validate and compare the accuracy of classifiers. RUSBoosted trees using data from waist and ankle activity monitors, with an accuracy of 99.1%, outperformed other classifiers in detecting three activities of interest.Clinical Relevance- One of the major aims of post-stroke rehabilitation is improving mobility, which may be facilitated by understanding the structure and pattern of everyday mobility through real-world, objective outcomes. Accurate activity identification, as shown in this pilot investigation, is an essential first step before developing objective outcomes for monitoring mobility and balance in everyday life of these individuals.

Falls are a significant health concern for individuals with traumatic brain injury (TBI). For developing effective preemptive strategies to reduce falls, it is essential to get an accurate and objective assessment of fall-risk. The current investigation evaluates the feasibility of a robotic, posturography-based fall-risk assessment to objectively quantify the risk of falls in individuals with TBI. Five individuals with chronic TBI (age: 56.2 ± 4.7 years, time since injury: 13.09±11.95 years) performed the fall-risk assessment on hunova- a commercial robotic platform for assessing and training balance. The unique assessment considers multifaceted fall-driving components, including static and dynamic balance, sit-to-stand, limits of stability, responses to perturbations, gait speed, and history of previous falls and provides a composite score for risk of falls, called silver index (SI), a number between 0 (no risk) and 100 (high risk) based on a machine learning-based predictive model. The SI score for individuals with TBI was 66±32.1 (min: 32, max: 100) - categorized as medium-to-high risk of falls. The construct validity of SI outcome was performed by evaluating its relationship with clinical outcomes of functional balance and mobility (Berg Balance Scale (BBS), Timed-Up and Go (TUG), and gait speed) as well as posturography outcomes (Center of Pressure (CoP) area and velocity). The bivariate Pearson correlation coefficient, although not statistically significant, suggested the presence of linear relationships (0.52 > r > 0.84) between SI and functional and posturography outcomes, supporting the construct validity of SI. A large sample is needed to further prove the validity of the SI outcome before it is used for meaningful interpretations of the risk of falls in individuals with TBI.Clinical Relevance- Clinical assessments of risk of falls are traditionally based on questionnaires that may lack objectivity, consistency, and accuracy. The current work tests the feasibility of using a robotic platform-based assessment to objectively quantify the risk of falls in individuals with TBI.

The objective of the current investigation was to evaluate the feasibility of a core-strengthening program delivered to a chronic stroke participant using a novel robotic device, AllCore360°, which targets trunk muscles through a systematic, consistent, high-intensity exercise. A 58-year old male with hemiplegia post stroke (time since injury: 18 years) was enrolled and performed 12-sessions of the core-strengthening program on AllCore360°. The participant completed a total of 142 360°-rotating-planks (called as 'spins') at four inclination angles, over 12 sessions. Assessments at baseline and follow up included posturography during quiet standing, electromyography (EMG) during AllCore360° spins, and assessments for trunk function (Trunk Impairment Scale (TIS)), balance (Berg Balance Scale (BBS) and mobility (Timed-Up and Go (TUG), 10-meter Walk test (10MWT), 6-minute Walk Test (6MWT)). Clinically meaningful improvements were observed in the TIS (73%), the BBS (45.2%), and the TUG test (22.7%). Medial-lateral Center of Pressure (MLCoP) data showed reduced RMS and range by 32.3% and 29.2%, respectively. EMG data from left and right rectus abdominis (RAB) muscles showed increased levels of activations for both inclination angles, 65° (LRAB: 74%, RRAB: 48.4%) and 55° (LRAB: 22.3%, RRAB: 28.7%). The participant rated the core-strengthening program 71 (scale: 0-126) on Physical ACtivity Enjoyment Scale at the follow up, showing a high level of satisfaction and engagement toward the training program. The preliminary results suggest that the novel robotic design and enhanced engagement of neuromuscular mechanisms features of AllCore360° core-strengthening program could facilitate improvements in trunk function, balance and mobility post stroke. A study with a large sample and an appropriate control group needs to be performed in the future.Clinical Relevance- The majority of clinical programs include core-stability exercises for improving trunk function. The current investigation presents a novel robotic-device based core-strengthening program that can provide systematic, consistent, and repetitive practice for optimal functional gains.

BACGROUND/UNASSIGNED:Interventions addressing balance dysfunction after traumatic brain injury (TBI) only target compensatory aspects and do not investigate perceptual mechanisms such as sensory acuity. OBJECTIVE:To evaluate the efficacy of a novel intervention that integrates sensory acuity with a perturbation-based approach for improving the perception and functional balance after TBI. METHODS:A two-group design was implemented to evaluate the effect of a novel, perturbation-based balance intervention. The intervention group (n = 5) performed the intervention with the sinusoidal (0.33, 0.5, and 1 Hz) perturbations to the base of support with amplitudes derived using our novel outcome of sensory acuity - perturbation perception threshold (PPT). The efficacy is evaluated using changes in PPT and functional outcomes (Berg Balance Scale (BBS), Timed-up and Go (TUG), 5-meter walk test (5MWT), and 10-meter walk test (10MWT)). RESULTS:There was a significant post-intervention change in PPT for 0.33 Hz (p = 0.021). Additionally, clinically and statistically significant improvements in TUG (p = 0.03), 5MWT (p = 0.05), and 10MWT (p = 0.04) were observed. CONCLUSIONS:This study provides preliminary efficacy of a novel, near-sensory balance intervention for individuals with TBI. The use of PPT is suggested for a comprehensive understanding and treatment of balance dysfunction. The promising results support the investigation in a larger cohort.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

OBJECTIVES:To derive accelerometer count thresholds for classifying time spent in sedentary, light intensity, and moderate-to-vigorous physical activity (MVPA) in manual wheelchair users (MWUs) with spinal cord injury (SCI). DESIGN:Participants completed 18 activities of daily living and exercises for 10 minutes each with a 3-minute break between activities while wearing a COSMED K4b2 portable metabolic cart and an ActiGraph activity monitor on the dominant wrist. A linear regression was computed between the wrist acceleration vector magnitude and SCI metabolic equivalent of task (MET) for 80% of the participants to obtain thresholds for classifying different activity intensities, and the obtained thresholds were tested for accuracy on the remaining 20% of participants. This cross-validation process was iterated for 1000 times to evaluate the stability of the thresholds on data corresponding to different proportions of sedentary, light intensity, and MVPA. MET values of 1.5 or lower were classified as sedentary behavior, MET values between 1.5 and 3 were classified as light intensity, and MET values of 3 or higher were classified as MVPA. The final thresholds were then validated on an out-of-sample independent dataset. PARTICIPANTS:MWUs (N=17) with SCI in the out-of-sample validation data set. INTERVENTIONS:Not applicable. SETTING:Research lab, community MAIN OUTCOME MEASURES: Accelerometer thresholds to classify sedentary, light intensity, and MVPA were obtained and their accuracy tested using cross-validation and an out-of-sample dataset. RESULTS:The threshold between sedentary and light intensity was 2057 counts-per-minute, and the threshold between light intensity and MVPA was 11,551 counts per minute. Based on the out-of-sample validation, the obtained thresholds had an overall accuracy of 85.6%, with a sensitivity and specificity of 95.3% and 97.4% for sedentary behavior, 87.8% and 84.5% for light intensity, 68.5% and 96.3% for MVPA, respectively. CONCLUSION:Accelerometer-based thresholds can be used to accurately identify sedentary behavior. However, thresholds may not provide accurate estimations of MVPA throughout the day when participants engage in more resistance-based activities.

STUDY DESIGN/METHODS:Cross-sectional validation study. OBJECTIVES/OBJECTIVE:To conduct a literature search for existing energy expenditure (EE) predictive algorithms using ActiGraph activity monitors for manual wheelchairs users (MWUs) with spinal cord injury (SCI), and evaluate their validity using an out-of-sample dataset. SETTING/METHODS:Research institution in Pittsburgh, USA. METHODS:A literature search resulted in five articles containing five sets of predictive equations using an ActiGraph activity monitor for MWUs with SCI. Out-of-sample data were collected from 29 MWUs with chronic SCI who were asked to follow an activity protocol while wearing an ActiGraph GT9X Link on the dominant wrist. They also wore a portable metabolic cart which provided the criterion measure for EE. The out-of-sample dataset was used to evaluate the validity of the five sets of EE predictive equations. RESULTS:. CONCLUSIONS:The existing EE predictive equations based on ActiGraph monitors for MWUs with SCI showed varied performance when compared with the criterion measure. Their accuracies may not be sufficient to support future clinical and research use. More work is needed to develop more accurate EE predictive equations for this population.

There is limited research on sensory acuity i.e., ability to perceive external perturbations via body-sway during standing in individuals with a traumatic brain injury (TBI). It is unclear whether sensory acuity diminishes after a TBI and if it is a contributing factor to balance dysfunction. The objective of this investigation is to first objectively quantify the sensory acuity in terms of perturbation perception threshold (PPT) and determine if it is related to functional outcomes of static and dynamic balance. Ten individuals with chronic TBI and 11 age-matched healthy controls (HC) performed PPT assessments at 0.33, 0.5, and 1 Hz horizontal perturbations to the base of support in the anterior-posterior direction, and a battery of functional assessments of static and dynamic balance and mobility [Berg balance scale (BBS), timed-up and go (TUG) and 5-m (5MWT) and 10-m walk test (10MWT)]. A psychophysical approach based on Single Interval Adjustment Matrix Protocol (SIAM), i.e., a yes-no task, was used to quantify the multi-sensory thresholds of perceived external perturbations to calculate PPT. A mixed-design analysis of variance (ANOVA) and post-hoc analyses were performed using independent and paired t-tests to evaluate within and between-group differences. Pearson correlation was computed to determine the relationship between the PPT and functional measures. The PPT values were significantly higher for the TBI group (0.33 Hz: 2.97 ± 1.0, 0.5 Hz: 2.39 ± 0.7, 1 Hz: 1.22 ± 0.4) compared to the HC group (0.33 Hz: 1.03 ± 0.6, 0.5 Hz: 0.89 ± 0.4, 1 Hz: 0.42 ± 0.2) for all three perturbation frequencies (p < 0.006 post Bonferroni correction). For the TBI group, the PPT for 1 Hz perturbations showed significant correlation with the functional measures of balance (BBS: r = -0.66, p = 0.037; TUG: r = 0.78, p = 0.008; 5MWT: r = 0.67, p = 0.034, 10MWT: r = 0.76, p = 0.012). These findings demonstrate that individuals with TBI have diminished sensory acuity during standing which may be linked to impaired balance function after TBI.