Work-related musculoskeletal disorders (WSMDs) are an important health concern in the workplace.  Accurately quantifying the level of physical exposures (i.e., kinematics and kinetics) is essential for risk assessments, developing and/or testing interventions, and improving estimates of exposure-response relationships.  Such exposures ideally should be quantified in situ, while workers interact with the actual work environment to complete their tasks.  However, in practice, relatively crude and/or time-consuming methods are often used, including self-reports, observational methods, and simple instrumentation, since directly assessing physical exposures is challenging in the workplace, and typically resource prohibitive. Inertial motion capture (IMC) and in-shoe pressure measurement (IPM) systems are emerging wearable technologies, and they can, respectively, facilitate monitoring of body kinematics and external forces on the body in the workplace.  Thus, this research examined the potential of such technologies in exposure assessments, and evaluated them in comparison to mature laboratory systems (i.e., optical motion capture system and force platform) or direct observation.  Performance of an IMC system was evaluated during several manual material handling (MMH) tasks, in terms of estimated body kinematics and kinetics at selected body parts.  A practical issue, regarding calibrating the IPM system in the field, was addressed by defining an ad hoc global coordinate system using a force platform.  Several regression models were developed for estimating center-of-pressure location and ground reaction forces.  Given that outputs from the IMC and the IPM systems are numerically fine-grained, but generally lack contextual information about a given job, task classification approaches were explored to automatically identify task types and their time proportions in a job. Overall, the outcomes from these studies demonstrated the potential of the IMC and the IPM systems for measuring physical exposures in the workplace.  However, estimation of physical exposures using these systems requires further improvements in some cases.  This research provided groundwork for future rapid and direct assessments of physical exposures in the workplace, and which needs to be expanded and validated in future efforts. / Ph. D.

Advances in wearable technologies are facilitating the understanding of human activities using first-person vision (FPV) for a wide range of assistive applications. In this thesis, we propose robust multiple motion features for human activity recognition from first-person videos. The proposed features encode discriminant characteristics from magnitude, direction and dynamics of motion estimated using optical flow. Moreover, we design novel virtual-inertial features from video, without using the actual inertial sensor, from the movement of intensity centroid across frames. Results on multiple datasets demonstrate that centroid-based inertial features improve the recognition performance of grid-based features. Moreover, we propose a multi-layer modelling framework that encodes hierarchical and temporal relationships among activities. The first layer operates on groups of features that effectively encode motion dynamics and temporal variations of intra-frame appearance descriptors of activities with a hierarchical topology. The second layer exploits the temporal context by weighting the outputs of the hierarchy during modelling. In addition, a post-decoding smoothing technique utilises decisions on past samples based on the confidence of the current sample. We validate the proposed framework with several classifiers, and the temporal modelling is shown to improve recognition performance. We also investigate the use of deep networks to simplify the feature engineering from firstperson videos. We propose a stacking of spectrograms to represent short-term global motions that contains a frequency-time representation of multiple motion components. This enables us to apply 2D convolutions to extract/learn motion features. We employ long short-term memory recurrent network to encode long-term temporal dependency among activities. Furthermore, we apply cross-domain knowledge transfer between inertial-based and vision-based approaches for egocentric activity recognition. We propose sparsity weighted combination of information from different motion modalities and/or streams. Results show that the proposed approach performs competitively with existing deep frameworks, moreover, with reduced complexity.

Due to the rapid development in electronics and radio communication systems, modern technologies are implemented to improve the safety and security of workplaces in order to save field service engineers lives and their health. In this thesis, an automated safety suit was implemented with integrated sensors for monitoring the safety of field service engineers. The basic idea of the prototype is to ensure safety for the field service engineers who are working in adverse environmental conditions. This safety suit includes embedded devices which can communicate with mobile devices and by means of that provides aid for the people working in several fields such as confined spaces, high altitudes etc. In this prototype, a different type of sensors are proposed for monitoring environmental and health conditions like temperature, CO gas levels, relative humidity, body temperature and heartbeat. A mobile application is proposed to monitor and control the automated safety suit, which also identifies the environmental changes and provide prompt alerts to the user. Keeping the usage of automated safety suit in mind, the system is designed in a user friendly manner and all the key elements are considered and implemented accordingly for the requirements of service engineers who are working in confined spaces and hazardous places.

BACKGROUND. Unhealthy lifestyle behaviours such as physical inactivity are global risk factors for chronic disease. Despite this, a substantial proportion of the UK population fail to achieve the recommended levels of physical activity. This may partly be because the health messages presently disseminated are not sufficiently potent to evoke behaviour change. There has been an exponential growth in the availability of digital health technologies within the consumer marketplace. This influx of technology has allowed people to self-monitor a plethora of health indices, such as their physical activity, in real-time. However, changing movement behaviours is difficult and often predicated on the assumption that individuals are willing to change their lifestyles today to reduce the risk of developing disease years or even decades later. One approach that may help overcome this challenge is to present physiological feedback in parallel with physical activity feedback. In combination, this approach may help people to observe the acute health benefits of being more physically active and subsequently translate that insight into a more physically active lifestyle. AIMS. Study One aimed to review existing studies employing fMRI to examine neurological responses to health messages pertaining to physical activity, sedentary behaviour, smoking, diet and alcohol consumption to assess the capacity for fMRI to assist in evaluating health behaviours. Study Two aimed to use fMRI to evaluate physical activity, sedentary behaviour and glucose feedback obtained through wearable digital health technologies and to explore associations between activated brain regions and subsequent changes in behaviour. Study Three aimed to explore engagement of people at risk of type 2 diabetes using digital health technologies to monitor physical activity and glucose levels. METHODS. Study One was a systematic review of published studies investigating health messages relating to physical activity, sedentary behaviour, diet, smoking or alcohol consumption using fMRI. Study Two asked adults aged 30-60 years to undergo fMRI whilst presented personalised feedback on their physical activity, sedentary behaviour and glucose levels, following a 14-day wear protocol of an accelerometer, inclinometer and flash glucose monitor. Study Three was a six-week, three-armed randomised feasibility trial for individuals at moderate-to-high risk of developing type 2 diabetes. The study used commercially available wearable physical activity (Fitbit Charge 2) and flash glucose (Freestyle Libre) technologies. Group 1 were offered glucose feedback for 4 weeks followed by glucose plus physical activity feedback for 2 weeks (G4GPA2). Group 2 were offered physical activity feedback for 4 weeks followed by glucose plus physical activity feedback for 2 weeks (PA4GPA2). Group 3 were offered glucose plus physical activity feedback for six weeks (GPA6). The primary outcome for the study was engagement, measured objectively by time spent on the Fitbit app, LibreLink app (companion app for the Freestyle Libre) as well as the frequency of scanning the Freestyle Libre and syncing the Fitbit. RESULTS. For Study One, 18 studies were included in the systematic review and of those, 15 examined neurological responses to smoking related health messages. The remaining three studies examined health messages about diet (k=2) and physical activity (k=1). Areas of the prefrontal cortex and amygdala were most commonly activated with increased activation of the ventromedial prefrontal cortex predicting subsequent behaviour (e.g. smoking cessation). Study Two identified that presenting people with personalised feedback relating to interstitial glucose levels resulted in significantly more brain activation when compared with feedback on personalised movement behaviours (P < .001). Activations within regions of the prefrontal cortex were significantly greater for glucose feedback compared with feedback on personalised movement behaviours. Activation in the subgyral area was correlated with moderate-to-vigorous physical activity at follow-up (r=.392, P=.043). In Study Three, time spent on the LibreLink app significantly reduced for G4GPA2 and GPA6 (week 1: 20.2±20 versus week 6: 9.4±14.6min/day, p=.007) and significantly fewer glucose scans were recorded (week 1: 9.2±5.1 versus week 6: 5.9±3.4 scans/day, p=.016). Similarly, Fitbit app usage significantly reduced (week 1: 7.1±3.8 versus week 6: 3.8±2.9min/day p=.003). The number of Fitbit syncs did not change significantly (week 1: 6.9±7.8 versus week 6: 6.5±10.2 syncs/day, p=.752). CONCLUSIONS. Study One highlighted the fact that thus far the field has focused on examining neurological responses to health messages using fMRI for smoking with important knowledge gaps in the neurological evaluation of health messages for other lifestyle behaviours. The prefrontal cortex and amygdala were most commonly activated in response to health messages. Using fMRI, Study Two was able to contribute to the knowledge gaps identified in Study One, with personalised glucose feedback resulting in a greater neurological response than personalised feedback on physical activity and sedentary behaviour. From this, Study Three found that individuals at risk of developing type 2 diabetes were able to engage with digital health technologies offering real-time feedback on behaviour and physiology, with engagement diminishing over time. Overall, this thesis demonstrates the potential for digital health technologies to play a key role in feedback paradigms relating to chronic disease prevention.

<div>Enhancing the capability to control a tremendous range of physical actuators and sensors, combined with wireless technology and the Internet of Things (IoT), apparel technologies play a significant role in supporting safe, comfortable and healthy living, observing each customer’s conditions. Since apparel technologies have advanced to enable humans to work as a team with the clothing they wear, the interaction between a human and apparel is further enhanced with the introduction of sensors, wireless network, and artificially intelligent techniques. A variety of wearable technologies have been developed and spread to meet the needs of customers, however, some wearable devices are considered as non-practical tech-oriented, not consumer-oriented.</div><div>The purpose of this research is to develop an apparel system which integrates intelligent autonomous agents, human-based sensors, wireless network protocol, mobile application management system and a zipper robot. This research is an augmentation to the existing research and literature, which are limited to the zipping and unzipping process without much built in intelligence. This research is to face the challenges of the elderly and people with self-care difficulties. The intent is to provide a scientific path for intelligent zipper robot systems with potential, not only to help people, but also to be commercialized.</div><div>The research develops an intelligent system to control of zippers fixed on garments, based on the profile and desire of the human. The theoretical and practical elements of developing small, integrated, intelligent zipper robots that interact with an application by using a lightweight MQTT protocol for application in the daily lives of diverse populations of people with physical challenges. The system functions as intelligent automatized garment to ensure users could positively utilize a zipper robot device to assist in putting on garments which also makes them feel comfortable wearing and interacting with the system. This research is an approach towards the “future of fashion”, and the goal is to incentivize and inspire others to develop new instances of wearable robots and sensors that help people with specific needs to live a better life.</div>

: Previous research has explored associations between accelerometry and Global Navigation Satellite System (GNSS) derived loads. However, to our knowledge, no study has investigated the relationship between these measures and a known distance. Thus, the current study aimed to assess and compare the ability of four accelerometry based metrics and GNSS to predict known distance completed using different movement constraints. A correlational design study was used to evaluate the association between the dependent and independent variables. A total of 30 physically active college students participated. Participants were asked to walk two different known distances (DIST) around a 2 m diameter circle (small circle) and a different distance around an 8 m diameter circle (large circle). Each distance completed around the small circle by one participant was completed around the large circle by a different participant. The same 30 distances were completed around each circle and ranged from 12.57 to 376.99 m. Acceleration data was collected via a tri-axial accelerometer sampling at 100 Hz. Accelerometry derived measures included the sum of the absolute values of acceleration (SUM), the square root of the sum of squared accelerations (MAG), Player Load (PL), and Impulse Load (IL). Distance (GNSSD) was measured from positional data collected using a triple GNSS unit sampling at 10 Hz. Separate simple linear regression models were created to assess the ability of each independent variable to predict DIST. The results indicate that all regression models performed well (R = 0.960-0.999, R = 0.922-0.999; RMSE = 0.047-0.242, < 0.001), while GNSSD (small circle, R = 0.999, R = 0.997, RMSE = 0.047 < 0.001; large circle, R = 0.999, R = 0.999, RMSE = 0.027, < 0.001) and the accelerometry derived metric MAG (small circle, R = 0.992, R = 0.983, RMSE = 0.112, < 0.001; large circle, R = 0.997, R = 0.995, RMSE = 0.064, < 0.001) performed best among all models. This research illustrates that both GNSS and accelerometry may be used to indicate total distance completed while walking.

Yes / Many children are not engaging in sufficient physical activity and there are substantial between-children physical activity inequalities. In addition to their primary role as educators, teachers are often regarded as being well-placed to make vital contributions to inclusive visions of physical activity promotion. With the dramatic increase in popularity of wearable technologies for physical activity promotion in recent years, there is a need to better understand teachers' perspectives about using such devices, and the data they produce, to support physical activity promotion in schools. Method: Semi-structured interviews were conducted with 26 UK-based primary school teachers, exploring their responses to children's physical activity data and their views about using wearable technologies during the school day. Interview discussions were facilitated by an elicitation technique whereby participants were presented with graphs illustrating children's in-school physical activity obtained from secondary wearable technology data. Interview transcripts were thematically analyzed. Results: Most teachers spoke positively about the use of wearable technologies specifically designed for school use, highlighting potential benefits and considerations. Many teachers were able to understand and critically interpret data showing unequal physical activity patterns both within-and between-schools. Being presented with the data prompted teachers to provide explanations about observable patterns, emotional reactions-particularly about inequalities-and express motivations to change the current situations in schools. Conclusion: These findings suggest that primary school teachers in the UK are open to integrating wearable technology for measuring children's physical activity into their practices and can interpret the data produced by such devices. Visual representations of physical activity elicited strong responses and thus could be used when working with teachers as an effective trigger to inform school practices and policies seeking to address in-school physical inactivity and inequalities.