21793 research outputs found
Sort by
Channel Surfing
No full textWhat is the nature of mind control? Does it exist? Sure, mind-altering drugs and hypnosis offer limited types of this control, but could media play a similar role? What role does media play in the structuring of our consciousness? This novella follows a consciousness that has become unmoored and loops through the world, its experience shaped by the media it previously had consumed and enjoyed. The main character, Lionel Grene, finds himself unable to escape the plot of a paranoid espionage novel that he has created.
Set in mid-1990s Ontario, Lionel
struggles to comprehend his situation, spiralling through the experience of the conspiracy filled plot of his spy novel. This plot is based in the true World War II history of Durham Region, Ontario. Places like Camp-X, Defence Industries Limited, and Camp 30 were real places. However, in Grene’s conditioned imagination, they become dens of plotting and manipulation. Then again, this might be historical fact too. Drawing on the novels of Thomas Pynchon as inspiration, a key aspect of this project is Grene’s slipping from fabulation (writing a novel) into confabulation (a memory error causing him to take himself as the protagonist of that novel). But beyond that, the project also interrogates the idea of media as subconscious programming
Investigating Zero-Shot Diagnostic Pathology in Vision-Language Models with Efficient Prompt Design
Get PDFVision-Language Models (VLMs) have emerged as powerful tools in computational pathology, offering the ability to perform zero-shot diagnostic inference on gigapixel whole slide images (WSIs). However, a core challenge remains: these models exhibit high sensitivity to the linguistic structure and specificity of prompts, which can significantly impact diagnostic accuracy, reproducibility, and clinical interpretability. This thesis systematically investigates the role of prompt engineering in enhancing the diagnostic performance of VLMs in histopathology. We propose a structured prompt engineering framework that modulates four critical dimensions: anatomical precision, information density, instructional framing, and output constraints, to evaluate their effect on model behavior. Using a clinically validated in-house dataset of 3,507 digestive system WSIs spanning multiple tissue types and pathological conditions, we conduct a comprehensive evaluation of four state of the art VLMs called Biomedical Contrastive Language-Image Pre-training (BioMedCLIP), Quilt-Net, Quilt-Large Language and Vision Assistant (Quilt-LLAVA), and Contrastive Learning from captions for Histopathology (CONCH). Our methodology includes a combination of quantitative assessments using Area Under the Curve (AUC) analysis , Receiver Operating Characteristic (ROC) and qualitative analyses to understand how prompt design influences diagnostic inference, interpretability, and generalization across tissues.
Our results demonstrate that prompt formulation significantly affects model performance across the full dataset. In particular, prompts that incorporate high anatomical specificity and clear instructional framing yield consistent improvements in classification accuracy across multiple tissue sites. The study further reveals that domain aligned prompting strategies are often more effective than increase in architectural complexity, highlighting the centrality of human and AI communication in medical vision-language tasks. In addition to empirical findings, we contribute actionable guidelines for implementing VLMs in clinical computational pathology workflows, emphasizing prompt standardization and interpretability. This work shifts the emphasis from purely architectural innovation to optimizing the language mediated interface between human expertise and AI systems, thereby enhancing both diagnostic performance and clinical utility in zero-shot medical image analysis
Microstructure and tribological performance of cu-based and self-lubricating coatings via low-temperature thermal spray
No full textThis thesis focuses on the development and evaluation of Cu-based coatings, specifically CuNiIn and self-lubricating CuNiIn/MoS₂ composites, fabricated using low-temperature thermal spray techniques. High Velocity Oxy-Fuel (HVOF), High Velocity Air-Fuel (HVAF), and Inner Diameter High Velocity Air-Fuel (ID-HVAF) processes were employed to deposit dense, low-porosity coatings aimed at improving tribological performance and reducing oxidation during fabrication.
Comprehensive microstructural and phase characterizations were performed using Field Emission Scanning Electron Microscopy (FE-SEM) to analyze surface and cross-sectional morphologies, X-ray Diffraction (XRD) to identify phase composition, and Raman Spectroscopy to confirm the integrity and distribution of MoS₂ in the composite coatings. Mechanical properties were evaluated through Vickers microhardness testing, and the tribological behavior of the coatings was systematically assessed using high-temperature reciprocating sliding wear tests at room temperature and 450 °C, simulating demanding service conditions.
The results revealed that coatings produced by HVAF and ID-HVAF exhibited significantly lower porosity and oxide content compared to HVOF, leading to improved mechanical integrity and wear resistance. The incorporation of MoS₂ as a solid lubricant within the CuNiIn matrix resulted in a substantial reduction in friction coefficient and wear rate, particularly at elevated temperatures, due to the formation of stable tribofilms and transfer layers at the sliding interface.
This study demonstrates the potential of HVAF-sprayed self-lubricating CuNiIn/MoS₂ composite coatings as high-performance solutions for aerospace and industrial components subjected to fretting and sliding wear. The findings contribute to the advancement of environmentally friendly and energy-efficient surface engineering technologies
Advancing eHMI for powered wheelchairs beyond safety and communication: a pilot study on enriching social interaction through a co-design approach
Get PDFBackground
Enhancing safety and communication while minimizing unwanted attention is key for wheelchair external Human-Machine Interfaces (eHMIs). This study aims to introduce an interface to enhance eHMIs for powered wheelchairs, improve external communication, and enhance positive social interactions in challenging urban situations.
Methods
A co-design approach was adopted, centering wheelchair users (WUs) in a two-step methodology. First, data were collected through a qualitative survey to define criteria, which informed themes for focus group discussions. These themes guided the ideation process. Eighteen participants, including WUs and experts in cognitive psychology, physiotherapy, and design, were involved. Concepts developed in ideation sessions were analyzed using the Analytic Hierarchy Process. A prototype was then developed to be assessed by both WUs and pedestrians through a structured questionnaire.
Results
According to the analysis, four themes were identified: I. Streamlined Information in Interaction, II. User-Centric Safety Feedback, III. Harmonious and Minimalist Interaction Design, and IV. Effortless Integration and Production. Regarding these themes, a table with design suggestions and implications was introduced. Ultimately, five interface concepts were proposed, with Concept 2, ‘WheelSafe Illumina’ (41.3%), and Concept 1, ‘WheelGlow Assist’ (28.1%) emerging as top priorities, both featuring a shell structure. Concept 2 was developed for prototyping. The feedback from the experiences of both WUs and pedestrians indicate that the proposed eHMI may enhance perceived communication and safety without drawing negative attention.
Conclusion
Integrating eHMI into a shell structure improves functional communication while also minimizing unwanted attention toward WUs—an often-overlooked issue in previous research that our co-design approach identified and effectively addressed
Platformization’s Elsewheres: Japanese Convenience Stores and the Platform Economy
Get PDFPlatformization’s elsewheres refers to other locations and places where platformization as a process takes place. This article focuses on the franchised Japanese convenience store as a particularly salient site from which to understand platformization in Japan. It is also crucial for thinking the platform economy historically and regionally within Asia where Japanese-style convenience stores abound, as well as globally given how Japan’s convenience stores were a model of the internet-connected mobile phone that in turn becomes a model for iPhone and Android smartphones. Focusing on the convenience store and its Japanese trajectory of development allows us to see the process of platformization of the franchised, networked, logistically-enabled convenience store from the 1970s to the present. The convenience store is, I argue, a crucial, if overlooked, site for platformization in Asia and beyond. It is also a key site for rethinking the most central of feelings to the platform economy: convenience
Gas discharge visualisation technology for psychological research applications: A systematic review
Get PDFGas discharge visualisation (GDV) is a nonintrusive technique that captures a person's physiological and psychoemotional status, including stress. This is through the electro-photonic emissions of fingertips placed on the surface of an impulse analyser. This study is to systematically review the GDV for psychological research applications through the PRISMA-NMA method. For the literature search, three databases, DOAJ, Scopus, and WOS, were consulted, using keyword terms, which allowed working with 20 articles published from 2002 to Feb 2025, demonstrating empirical evidence of the impact of GDV for psychological applications
L’utilisation d’un modèle meaning-making en art-thérapie pour les enfants atteints de douleur chronique
Get PDFLa douleur chronique juvénile est une expérience complexe ayant de nombreux impacts sur les plans physique, psychologique et psychosocial. Touchant près d’un enfant sur cinq au Canada, les deux tiers de ces jeunes continueront de souffrir à l’âge adulte (SKIP, 2021). En plus, seulement 5 à 15 % de ces jeunes reçoivent réellement des soins cliniques, alors que la littérature confirme qu’une prise en charge précoce et multidisciplinaire contribue à une meilleure qualité de vie pour ces enfants (Gouvernement du Canada, 2023). De plus, il existe très peu de ressources permettant d’offrir un soutien psychosocial, un besoin pourtant grandement exprimé. L’art-thérapie est une modalité thérapeutique flexible, alliant forces individuelles et créativité afin de pallier les manques psychosociaux nommés par la littérature. Ainsi, le but de ce travail de recherche est d’élaborer un cadre d’intervention en art-thérapie pour des jeunes vivant avec de la douleur chronique. Considérant que la douleur chronique peut engendrer de nombreuses pertes, cette recherche est informée par une théorie meaning-making, un concept récurrent dans les thérapies du deuil. La méthodologie choisie considère les premières étapes d’une recherche-intervention telle que conceptualisée par Fraser et Galinski (2009 ; 2010), soit la définition du problème, l’analyse de la littérature et la proposition d’une théorie du changement. Toute l’information est collectée et organisée de manière à déterminer une structure et les processus essentiels pour créer un cadre d’intervention. Bien que ce programme ne soit pas évalué et peaufiné dans un contexte pratique, il sert de guide et de plaidoyer pour quiconque voulant établir un programme ou intervenir auprès de cette population
Pedestrians’ social empathy and interaction with wheelchair users: The impact of user gestures and mobility aid design in a pilot study
Get PDFWheelchair users (WUs) experience various accessibility challenges in public spaces, which may lead them to seek assistance from pedestrians in difficult situations. In this context, understanding the factors influencing pedestrians' empathy and interaction with WUs facilitates their social interactions in challenging urban situations. This study examines how WUs’ body gestures and wheelchair design characteristics (WDCs) impact pedestrian perception and interaction. A pilot cross-sectional study was conducted with 52 participants in two phases: (1) a questionnaire assessing willingness to engage with WUs exhibiting independent or help-seeking gestures, and (2) evaluating four wheelchair types—from conventional to advanced—using key semantic descriptors of appearance and social perception. Findings revealed no significant relationship between age, gender, and willingness to interact across the two gesture conditions (p > 0.05), except for a significant association between age and willingness to interact with users of advanced powered wheelchairs in the help-seeking gesture condition (p = 0.027). Also, pedestrians' willingness to interact was significantly higher when WUs exhibited help-seeking gestures compared to independence gestures (p < 0.001). WDCs influenced pedestrian perceptions more strongly when WUs displayed independence (86.3 %) than help-seeking gestures (50 %). Moreover, analysis of semantic evaluations revealed distinct perceptual dimensions for advanced manual and powered wheelchairs, with three principal components identified for each, offering valuable insights for developing wheelchairs with greater social polish. This study highlights that both WDC and user gestures significantly affect pedestrian interaction, with the masking effect of help gestures on WDCs being a key finding. Additionally, advanced WDCs signify WUs' independence, helping reduce negative social stereotypes among pedestrians
Informative Machine Learning Model Explanation Techniques
Get PDFExplainable AI (XAI) is an emerging field focused on providing human-interpretable insights into complex and often black-box machine learning (ML) models. Shapley value attribution (SVA) is an increasingly popular XAI method that quantifies the contribution of each feature to a model’s behavior, which can be either an individual prediction (local SVAs) or a performance metric (global SVAs). However, recent research has highlighted several limitations in existing SVA methods, leading to biased or incorrect explanations that fail to capture the true relationships between features and model behaviors. What's worse, these explanations are vulnerable to adversarial manipulation.
Additionally, global SVAs, while widely used in applied studies to gain insights into underlying information systems, face challenges when applied to ML models trained on imbalanced datasets, such as those used in fraud detection or disease prediction. In these scenarios, global SVAs can yield misleading or unstable explanations.
This thesis aims to address these challenges and improve the reliability and informativeness of SVA explanations. It makes three key contributions: 1) Proposing a novel error analysis framework that comprehensively examines the underlying sources of bias in existing SVA methods; 2) Introducing a series of refinement methods that significantly enhance the informativeness of SVA explanations, as well as their robustness against adversarial attacks; 3) Developing a standardization method for evaluating global model behaviors on imbalanced datasets, advancing the development of an explainable model monitoring system. Our experiments demonstrate that these methods substantially improve the ability of SVAs to uncover informative patterns in model behaviors, making them valuable tools for knowledge discovery, model debugging, and performance monitoring
What the Body Knows: A Code for Living
Get PDFThis research-creation MA thesis asks not just how our bodies are relational but how the process of embodiment is relational, and shows why this is important to collective flourishing. Prioritizing this alongside the unfolding climate crisis, the primary question in this thesis explores how our body’s intelligence intersects with planetary wellness, amplifying innate principles of reciprocity that exist throughout our bodies and the wider systems of life on Earth. Presenting healing modalities and strategies that emphasize the body's intelligence and centre concepts of interconnection, this research unites personal ritual, ceremony, dream and ancestor work, somatic therapy and plant medicine with art-making, scholarly research and writing.
The resulting large-scale textile works, comprised of botanical cyanotypes, hand-embroideries and bilateral drawings on cloth, work together with a written thesis to critically examine the fascia of connectivity between individual, community and planetary wellness. Based in the offer of deep medicine to repair relationships “damaged through systems of domination”, this thesis juxtaposes Western medical frameworks around healing against personal experience and the wisdom of traditional and Indigenous cultures. This concludes with redefining the human body and health to encompass the living-systems theory of Healing Justice, where mutual flourishing forms the heart of personal and collective wellness, and showing that our body’s way-of-knowing is inherently linked to planetary health