Browsing by Author "Lim, Sol Ie"

Now showing 1 - 8 of 8
  • Thumbnail Image
    Armband EMG-based Lifting Detection and Load Classification Algorithms using Static and Dynamic Lifting Trials
    Taori, Sakshi Pranay (Virginia Tech, 2023-06-08)
    The high prevalence of work-related musculoskeletal disorders in occupational settings necessitates the development of economic, accurate, and convenient methods for quantifying biomechanical risk exposures. In terms of lifting, the occupational work environment does not provide resources for recording the start and end times of lifting tasks performed by individual workers. As a result, automatic detection of lift starts and ends is required for practical purposes. Occupational lifting styles vary depending on the asymmetry angle, which is the degree of shoulder or trunk rotation required by the lifting task. Predictive or machine learning (ML) algorithms have been increasingly used in the ergonomics field to identify occupational risk factors, such as lifting loads. However, such algorithms are often developed and validated using the dataset collected from the same lab-based experimental set-up, which limits their external validity. The recent development of wearable armbands with surface electromyography (sEMG) electrodes provides a low-cost, wireless, and non-invasive way to collect EMG data beyond laboratory settings. Despite their tremendous potential for field-based workload estimation, these armbands have not been widely implemented yet in automated lift detection and occupational workload estimation. The objective of this study was to evaluate the performance of machine learning (ML) algorithms in the automatic detection of lifts and classification of hand loads during manual lifting tasks from the data acquired by a wearable armband sensor with eight surface electromyography (sEMG) electrodes. Twelve healthy participants (six male and six female) performed repetitive symmetric (S), asymmetric (A), and free dynamic (F) lifts with three different hand-load levels (5 lb, 10 lb and 15 lb) at two origin (24" and 36") and two destination heights (6" and 36"). Three ML algorithms were utilized: Random Forest (RF), Support Vector Machines (SVM) and Gaussian Naïve Bayes (GNB). For lift detection, a subset of four participants was analyzed as a preliminary investigation. RF showed the best performance with the mean start and end errors of 0.53 ± 0.25 seconds and 0.76 ± 0.28 seconds, respectively. The accuracy score of 84.3 ± 3.3% was reported for lift start and 83.3 ± 9.9% for lift end. For hand-load classification, prediction models were developed using four different lifting datasets (S, A, S+A, and F) and were cross-validated using F as the test dataset. Mean classification accuracy was significantly lower in models developed with the S dataset (78.8 ± 7.3%) compared to A (83.3 ± 7.2%), S+A (82.1 ± 7.3%), and F (83.4 ± 8.1%). Overall, findings indicate that the implementation of ML algorithms with wearable EMG armbands for automatic lift detection in occupational settings can be promising. In hand-load classification, models developed with only controlled symmetric lifts were less accurate in predicting loads of more dynamic, unconstrained lifts, which is common in real-world settings. However, since both A and S+A demonstrated equivalent model accuracy with F, EMG armbands possess strong potential for estimating the hand loads of free-dynamic lifts using constrained lift trials involving asymmetric lifts.
  • Thumbnail Image
    The Effect of Mental Fatigue on Risk of Falling
    Abuhaija, Laith Ayman (Virginia Tech, 2022-01-18)
    Slips, trips, and falls are the costliest source of disabling injuries in the workplace, costing $18.6 billion annually. The purpose of this study was to investigate the effects of mental fatigue on gait variables associated with the risk of slipping and tripping. The study also investigated the efficacy of a 10-minute rest break in mitigating the effect of mental fatigue on those variables. Twenty healthy young adults (10 males and 10 females) participated and completed two experimental sessions. The order of sessions was counter-balanced for each participant. During the mental fatigue session, participants completed a computerized mentally fatiguing task for 90 minutes and performed a set of gait trials every 15 minutes throughout the task. During the control session, participants watched an emotionally neutral documentary in place of the mentally fatiguing task. After 90 minutes of the task or documentary, participants took a 10- minute break and then completed one last set of gait trials. Risk of slipping was inferred from the required coefficient of friction, heel contact velocity, and heel contact angle. Risk of tripping was inferred from minimum toe clearance and obstacle clearance. The results showed no increase in slip or trip risk. Rest breaks appeared to decrease levels of self-reported mental fatigue. However, they did not appear to have any mitigating effect on any of the gait variables that were measured.
  • Thumbnail Image
    The Effects of a Humanoid Robot's Non-lexical Vocalization on Emotion Recognition and Robot Perception
    Liu, Xiaozhen (Virginia Tech, 2023-06-30)
    As robots have become more pervasive in our everyday life, social aspects of robots have attracted researchers' attention. Because emotions play a key role in social interactions, research has been conducted on conveying emotions via speech, whereas little research has focused on the effects of non-speech sounds on users' robot perception. We conducted a within-subjects exploratory study with 40 young adults to investigate the effects of non-speech sounds (regular voice, characterized voice, musical sound, and no sound) and basic emotions (anger, fear, happiness, sadness, and surprise) on user perception. While listening to the fairytale with the participant, a humanoid robot (Pepper) responded to the story with a recorded emotional sound with a gesture. Participants showed significantly higher emotion recognition accuracy from the regular voice than from other sounds. The confusion matrix showed that happiness and sadness had the highest emotion recognition accuracy, which aligns with the previous research. Regular voice also induced higher trust, naturalness, and preference compared to other sounds. Interestingly, musical sound mostly showed lower perceptions than no sound. A further exploratory study was conducted with an additional 49 young people to investigate the effect of regular non-verbal voices (female voices and male voices) and basic emotions (happiness, sadness, anger, and relief) on user perception. We also further explored the impact of participants' gender on emotion and social perception toward robot Pepper. While listening to a fairy tale with the participants, a humanoid robot (Pepper) responded to the story with gestures and emotional voices. Participants showed significantly higher emotion recognition accuracy and social perception from the voice + Gesture condition than Gesture only conditions. The confusion matrix showed that happiness and sadness had the highest emotion recognition accuracy, which aligns with the previous research. Interestingly, participants felt more discomfort and anthropomorphism in male voices compared to female voices. Male participants were more likely to feel uncomfortable when interacting with Pepper. In contrast, female participants were more likely to feel warm. However, the gender of the robot voice or the gender of the participant did not affect the accuracy of emotion recognition. Results are discussed with social robot design guidelines for emotional cues and future research directions.
  • Thumbnail Image
    Enhancing Online Yoga Instruction: Evaluating the Effectiveness of Visual Augmentations for Performance Assessment
    Gopal, Ajit Ayyadurai (Virginia Tech, 2024-10-23)
    Yoga is a mind-body practice known for its substantial psychological and physiological benefit, contributing to a healthy lifestyle. However, without professional guidance, individuals may experience reduced performance and increased risk of injury. While online yoga classes on platforms like Zoom have grown in popularity, tools to support instructors in accurately assessing and monitoring student performance remain insufficient. For certain populations, this lack of real-time professional guidance poses safety risks and limits the effectiveness of the practice. This study examined the effectiveness of using computer-vision-based visual augmentations in enhancing instructors' ability to assess student performance and ensure safety. Specifically, we investigated the effectiveness of various visual augmentations in aiding instructors' visual search for unstable or unsafe poses. Eleven certified yoga instructors (8 female, 3 male), each holding 200 to 500 RYT certifications, participated in the study. Instructors completed eight trials assessing 12 yoga poses using four different visual augmentations—Raw Video, Skeleton (joint locations overlay), Contour (participant outlines), and Contour + Skeleton—across two camera views (Single vs. Multiple Views). During each trial, eye-tracking data was collected as instructors identified potentially unstable (unsafe) poses, and they subsequently completed a usability questionnaire and NASA - TLX rating. Upon finishing all trials, instructors provided overall feedback on the usability of the visual augmentations and camera views Instructors showed no significant difference in their assessment performance across different visual augmentations and camera views. The Skeleton augmentation led to increased cognitive workload, as indicated by larger pupil diameters. The Contour alone augmentation was less effective for visual search based on the usability ratings, and combining Contour with Skeleton did not offer notable improvements. Simpler visualizations, such as Raw and Skeleton, received higher usability ratings, and instructors preferred Single View layouts over Multiple Views for their ease of use and lower cognitive demand. In conclusion, while Skeleton augmentation increased cognitive load, it did not significantly enhance visual search performance. Future research should explore alternative visual augmentation techniques and configurations to better assist instructors on performance assessment which increases overall performance while not substantially increasing cognitive workload.
  • Thumbnail Image
    Evaluating Alternative Inertial Measurement Unit Locations on the Body for Slip Recovery Measures
    Morris, Michelle Ann (Virginia Tech, 2024-04-03)
    Slips are a leading cause of injury among older adults. Specific slip recovery measures, including slip distance and peak slip speed, have been shown to increase significantly among fallers as compared to non-fallers. Often, slipping kinematics are measured using optoelectronic motion capture (OMC), requiring a laboratory setting and limiting data collection to experimentally-controlled conditions. Inertial measurement units (IMUs) show promise as a portable and wearable form of motion capture. This study had two objectives. First, we investigated whether foot and ankle IMU-derived slip recovery measures could be considered equivalent to the same OMC-derived measures. Second, we investigated if both participant-placed and researcher-placed IMU-derived slip recovery measures could be considered equivalent to the same OMC-derived measures. 30 older adults (ages 65-80) were exposed to a slip while wearing both IMUs and OMC markers. Slip distance and peak slip speed were measured by both systems and compared. Equivalence testing (α = 0.05) showed that IMUs placed on the foot and the ankle were equivalent to OMC in measuring these slip recovery measures. Furthermore, it was shown that researcher and participant-placed IMUs were equivalent (α = 0.05) to OMC in measuring these slip recovery measures. These results confirm that IMUs can be a viable substitute for OMC and have the potential to expand data capture to a real-world environment.
  • Thumbnail Image
    Investigating the Relationship Between Objective and Subjective Measures of Physical Demand During Passive Exoskeleton Use
    Kelley, Sydney Aelish (Virginia Tech, 2023-10-24)
    Passive exoskeletons hold promise in reducing the risk of work-related musculoskeletal disorders, however further research is essential before widespread adoption can occur. This study explores the feasibility of using subjective measures of physical demand in place of costly and less practical objective measures. Normalized electromyography (nEMG) data and ratings of perceived exertion (RPE) were collected from seven different studies conducted by the Occupational Ergonomics and Biomechanics Lab (OEB lab). Employing a repeated measures three-way ANOVA, we assessed the influence of nEMG, gender, and exoskeleton type on RPE. Additionally, mean nEMG and RPE from seven passive exoskeleton-based studies conducted outside the OEB lab were assessed in order to determine if the findings from the OEB lab existed across other research environments. The results demonstrated a general positive linear trend between nEMG and RPE for both the individual and mean results. Substantial inconsistencies emerged when considering the influence of gender, exoskeleton type, and task conditions on the relationship between nEMG and RPE. These discrepancies underscore the need for more in-depth research into this topic, specifically investigating the effects of gender and exoskeleton design.
  • Thumbnail Image
    Non-Treadmill Trip Training – Laboratory Efficacy, Validation of Inertial Measurement Units, and Tripping Kinematics in the Real World
    Lee, Youngjae (Virginia Tech, 2024-06-05)
    Trip-induced falls are a leading cause of injuries among adults aged 65 years or older. Perturbation-based balance training (PBT) has emerged as an exercise-based fall prevention intervention and shown efficacy in reducing the risk of trip-induced falls. The broad goal of my PhD research was to advance the application of this so-called trip training through three studies designed to address existing knowledge gaps. First, trip training is commonly conducted with the aid of costly specialized treadmills to induce trip-like perturbations. An alternative version of trip training that eliminates the need for a treadmill would enhance training feasibility and enable wider adoption. The goal of the first study was to compare the effects of non-treadmill training (NT), treadmill training (TT), and a control (i.e., no training) on reactive balance after laboratory-induced trips among community-dwelling older adults. After three weeks of the assigned intervention, participants were exposed to two laboratory-induced trips while walking. Results showed different beneficial effects of NT and TT. For example, NT may be more beneficial in improving recovery step kinematics, while TT may be more beneficial in improving trunk kinematics, compared to the control. While the first study showed the effects of PBT on laboratory-induced trips, little is known about how such training affects responses to real-world trips. Responses to real-world trips may be captured using wearable inertial measurement units (IMUs), yet IMUs have not been adequately validated for this use. Therefore, the goal of the second study was to investigate the concurrent validity of IMU-based trunk kinematics against the gold standard optical motion capture (OMC)-based trunk kinematics after overground trips among community-dwelling older adults. During two laboratory-induced trips, participants wore two IMUs placed on the sternum and shoulder, and OMC markers placed at anatomical landmarks of the trunk segment. Results showed that IMU-based trunk kinematics differed between falls and recoveries after overground trips, and exhibited at least good correlation (Pearson's correlation coefficient, r > 0.5) with the gold standard OMC-based trunk kinematics. The goal of the third study was then to explore differences in tripping kinematics between the laboratory and real world using wearable IMUs among community-dwelling older adults. Participants were asked to wear three IMUs (for sternum and both feet) and a voice recorder to capture their responses to real-world losses of balance (LOBs) during their daily activities for three weeks. Results showed a higher variance in laboratory-induced trips than real-world trips, and the study demonstrated the feasibility of using IMUs and a voice recorder to understand the underlying mechanisms and context of real-world LOBs. Overall, this work was innovative by evaluating a non-treadmill version of trip training, establishing the validity of IMUs in capturing kinematic responses after overground trips, and applying IMUs and a voice recorder to assess tripping kinematics in the real world. The results from this work will advance the use of PBT to reduce the prevalence of trip-induced falls and to investigate the real-world effects of such trip training in future studies.
  • Thumbnail Image
    Quantifying the Reliability of Performance Time and User Perceptions Obtained from Passive Exoskeleton Evaluations
    Noll, Alexander Baldrich Benoni (Virginia Tech, 2024-08-16)
    Work-related musculoskeletal disorders (WMSDs) cost US industries billions annually and reduce quality of life for those afflicted. Passive exoskeletons (EXOs) have emerged as a potential intervention to reduce worker exposures to WMSD risk factors. As EXO adoption is rising, EXO manufacturers are designing and producing new EXOs in accordance with growing demand. However, there are no standardized EXO evaluation protocols and EXO use recommendations, due in part to insufficient information on the reliability of EXO evaluation measures. The purpose of this thesis was to quantify the reliability of common EXO evaluation measures, using both traditional approaches a more advanced statistical approach (i.e., Generalizability Theory), while also identifying potential effects of EXO type, work task, and individual differences. This work used data from a recently completed EXO evaluation study, conducted in Virginia Tech's Occupational Ergonomics and Biomechanics Lab. Forty-two total participants completed simulated occupational tasks, in two separate experimental sessions on different days, while using an arm-support EXO (ASE) and a back-support EXO (BSE). Several outcome measures reached excellent within-session reliability within four trials for many tasks considered. Between-session reliability levels were lower than within-session levels, with outcome measures reaching moderate-to-good reliability for most tasks. Interindividual differences accounted for the largest proportion of variance for measurement reliability, followed by the experimental session. For all tasks, outcome measures reached excellent dependability levels, with many achieving excellent levels within five total trials. Inconsistencies observed in between-session reliability levels and dependability levels suggest that additional training and EXO familiarity may affect measurement reliability of outcome measures differently for some tasks, unique to each EXO type. These discrepancies emphasize the importance for additional research into this topic. Overall, the current findings indicate that many of the commonly used EXO evaluation measures are reliable and dependable within five trials and one experimental session, providing a potential foundation for standardized EXO assessment protocols.