No description available.

Trust is explored in this thesis through the analysis of the robustness of wearable device based artificial intelligence based models to changes in data acquisition. Specifically changes in wearable device hardware and different recording sessions are explored. Three human activity recognition models are used as a vehicle to explore this: Model A which is trained using accelerometer signals recorded by a wearable sensor referred to as Astroskin, Model H which is trained using accelerometer signals from a wearable sensor referred to as the BioHarness and Model A Type 1 which was trained on Astroskin accelerometer signals that was recorded on the first session of the experimental protocol. On a test set recorded by Astroskin Model A had a 99.07% accuracy. However on a test set recorded by the BioHarness Model A had a 65.74% accuracy. On a test set recorded by BioHarness Model H had a 95.37% accuracy. However on a test set recorded by Astroskin Model H had a 29.63% accuracy. Model A Type 1 an average accuracy of 99.57% on data recorded by the same wearable sensor and same session. An average accuracy of 50.95% was obtained on a test set that was recorded by the same wearable sensor but by a different session. An average accuracy of 41.31% was obtained on data that was recorded by a different wearable sensor and same session. An average accuracy of 19.28% was obtained on data that was recorded by a different wearable sensor and different session. An out of domain discriminator for Model A Type 1 was also implemented. The out of domain discriminator was able to differentiate between the data that trained Model A Type 1 and other types (data recorded by a different wearable devices/different sessions) with an accuracy of 97.60%. / Thesis / Master of Applied Science (MASc) / The trustworthiness of artificial intelligence must be explored before society can fully reap its benefits. The element of trust that is explored in this thesis is the robustness of wearable device based artificial intelligence models to changes in data acquisition. The specific changes that are explored are changes in the wearable device used to record the input data as well as input data from different recording sessions. Using human activity recognition models as a vehicle, the results show that performance degradation occurs when the wearable device is changed and when data comes from a different recording session. An out of domain discriminator is developed to alert users when a potential performance degradation can occur.

No description available.

No description available.

No description available.

This project aims to explore how interactive technological artefacts can support young adults with sensory overload in their experiences. It intends to investigate the potential of carefully designing technologies tailored to the user's sensory needs of managing and reflecting on the sensory experience, and understanding their patterns. The project adopted an iterative design process and applied several methods with an emphasis on designing meaningful user experiences. The project conducted user research with people experiencing sensory overload to understand the experiential aspects of trying to manage the negative experiences it entails. The final design consists of two prototypes. The first one was a Role Prototype created with Figma and took the form of a mHealth app powered by Artificial Intelligence for detecting patterns in the user’s data and documenting sensory experiences through journaling. The second one is an Implementation prototype in the form of a tangible electronic prototype created by using Arduino Nano BLE 33 Sense and sensors for recognising biometric data and environmental cues. Additionally, technological exploration was undertaken through sketching in hardware. This led to delving into how Machine Learning and gas sensors can be combined in order to create a scent-detecting sensor. The results suggest that the design concept provides the user with valuable tools to manage their overwhelming sensory experiences. Lastly, the results indicate that a sensory experience prediction component would be a valuable feature for people who need to manage sensory overload.

Introduction: Wearable technologies play an important contemporary role in the measurement of physical activity (PA) and promotion of human health across the lifespan, including for young people (i.e., children, adolescents, and young adults). As new objective wearable technologies continue to develop, standardized approaches to documenting validation parameters (i.e., measures of accuracy, precision, and bias) are needed to ensure confidence and comparability in step-defined PA. Purpose: To produce validity parameters for step counting wearable technologies during treadmill walking in young people 6-20 years of age Methods: 120 participants completed 5-minute treadmill bouts from13.4 to 134.1 m·min-1. Participants wore eight technologies (two at the arm/wrist, four at the waist, one on the thigh, and one on the ankle) while steps were directly observed. Speed, wear location, and age -specific measures of accuracy (mean absolute percent error; MAPE), precision (correlation coefficient, standard deviation; SD, coefficient of variation; CoV), and bias (percent error; PE) were computed and cataloged. Results: Speed and wear location had a significant effect on accuracy and bias measures for wearable technologies (pConclusion: While the analyses indicate the significance of speed and wear location on wearable technology performance, the useful and comprehensive validity reference values cataloged herein will help optimize measurement of PA in youth. Future research should continue to rigorously validate new wearable technologies as they are developed, and also extend these standardized reference values developed in the laboratory to the free-living environment.

Benefit from the recent advancements of artificial intelligence (AI) methods, industrial automation has replaced human labors in many tasks. However, humans are still placed in the central role when visual searching tasks are highly involved for manufacturing decision-making. For example, highly customized products fabricated by additive manufacturing processes have posed significant challenges to AI methods in terms of their performance and generalizability. As a result, in practice, human visual searching tasks are still widely involved in manufacturing contexts (e.g., human resource management, quality inspection, etc.) based on various visualization techniques. Quantitatively modeling the visual searching behaviors and performance will not only contribute to the understanding of decision-making process in a visualization system, but also advance AI methods by incubating them with human expertise. In general, visual searching can be quantitatively understood from multiple scales, namely, 1) the population scale to treat individuals equally and model the general relationship between individual's physiological signals with visual searching decisions; 2) the individual scale to model the relationship between individual differences and visual searching decisions; and 3) the attention scale to model the relationship between individuals' attention in visual searching and visual searching decisions. The advancements of wearable sensing techniques enable such multiscale quantitative analytics of human visual searching performance. For example, by equipping human users with electroencephalogram (EEG) device, eye tracker, and logging system, the multiscale quantitative relationships among human physiological signals, behaviors and performance can be readily established. This dissertation attempts to quantify visual searching process from multiple scales by proposing (1) a data-fusion method to model the quantitative relationship between physiological signals and human's perceived task complexities (population scale, Chapter 2); (2) a recommender system to quantify and decompose the individual differences into explicit and implicit differences via personalized recommender system-based sensor analytics (individual scale, Chapter 3); and (3) a visual language processing modeling framework to identify and correlate visual cues (i.e., identified from fixations) with humans' quality inspection decisions in human visual searching tasks (attention scale, Chapter 4). Finally, Chapter 5 summarizes the contributions and proposes future research directions. The proposed methodologies can be readily extended to other applications and research studies to support multi-scale quantitative analytics. Besides, the quantitative understanding of human visual searching behaviors performance can also generate insights to further incubate AI methods with human expertise. Merits of the proposed methodologies are demonstrated in a visualization evaluation user study, and a cognitive hacking user study. Detailed notes to guide the implementation and deployment are provided for practitioners and researchers in each chapter. / Doctor of Philosophy / Existing industrial automation is limited by the performance and generalizability of artificial intelligence (AI) methods. Therefore, various human visual searching tasks are still widely involved in manufacturing contexts based on many visualization techniques, e.g., to searching for specific information, and to make decisions based on sequentially gathered information. Quantitatively modeling the visual searching performance will not only contribute to the understanding of human behaviors in a visualization system, but also advance the AI methods by incubating them with human expertise. In this dissertation, visual searching performance is characterized from multiple scales, namely, 1) the population scale to understand the visual searching performance in regardless of individual differences; 2) the individual scale to model the performance by quantifying individual differences; and 3) the attention scale to quantify the human visual searching-based decision-making process. Thanks to the advancements in wearable sensing techniques, this dissertation attempts to quantify visual searching process from multiple scales by proposing (1) a data-fusion method to model the quantitative relationship between physiological signals and human's perceived task complexities (population scale, Chapter 2); (2) a recommender system to suggest the best visualization design to the right person at the right time via sensor analytics (individual scale, Chapter 3); and (3) a visual language processing modeling framework to model humans' quality inspection decisions (attention scale, Chapter 4). Finally, Chapter 5 summarizes the contributions and proposes future research directions. Merits of the proposed methodologies are demonstrated in a visualization evaluation user study, and a cognitive hacking user study. The proposed methodologies can be readily extended to other applications and research studies to support multi-scale quantitative analytics.

This thesis presents a method for estimating the positions of human limbs during motion that can be applied to wearable, textile-based sensors. The method was validated for the elbow and shoulder joints with data from two garments with resistive, thread-based sensors sewn into the garments at multiple locations. The proposed method was able to estimate the elbow joint position with an average error of 2.2 degrees. The method also produced an average difference in Euclidean distance of 3.7 degrees for the estimated shoulder joint position using data from nine sensors placed around the subject's shoulder. The most accurate combination of sensors on the shoulder garment was found to produce an average difference in distance of 3.4 degrees and used only six sensors. The characteristics of the resistive, thread-based sensor used to validate the method are also detailed as some of their behaviors proved to affect the accuracy of the method negatively. / Master of Science / Human motion capture systems gather data on the position of the human body during motion. The data is then used to recreate and analyze the motion digitally. There is a need for motion capture devices capable of measuring long-term data on human motion, especially in physical therapy. However, the currently available motion capture systems have limitations that make long-term or daily use either impossible or uncomfortable. This thesis presents a method that uses data from wearable, textile-based sensors to estimate the positions of human limbs during motion. Two garments were used to validate the method on the elbow and shoulder joints. The proposed method was able to measure the elbow and shoulder joints with an average accuracy that is within the acceptable range for clinical settings.

This thesis presents the design and analysis of a vest-worn alerting system with the purpose of warning those who wear it. The goal of this thesis has been shaped around roadside construction workers, people who endeavor to build and maintain the roadways in highly noisy and dangerous environments. Our goal is to determine what type of alerting method best interrupts the worker, allowing him or her to know that danger approaches. Multiple alerting methods will be compared via reaction time and user data derived through critical incidents from the Critical Incident Technique (CIT) and notification-focused questionnaires. All testing is done in a simulated noisy environment. Each of these alerting methods involve combinations of auditory, visual, or haptic components. At the end of this thesis, the different alerting methods will be compared and a mode will be suggested for wearable notification activities. The work proposed in this thesis focuses mostly on vest design and alert testing for construction-based scenarios, but the work can be extended to police and highway worker scenarios. / Master of Science