41530 research outputs found
Sort by
Towards an Ecofeminist Pedagogy: The Environmental Ethics of AI
No full textPrevious scholarship has outlined some of the ethical considerations regarding the use of artificial intelligence (AI) in the classroom and beyond, such as the potentials for plagiarism, information bias, disinformation, or even violations of user privacy. Though research suggests that machine learning is a major and ever-increasing contributor to carbon emissions, there has been little discussion of how we teach about the ethics of AI from an environmental perspective. Therefore, this critical commentary first outlines how an ecofeminist pedagogy can help students consider the environmental and ecological effects of new technologies, including AI. At its core, an ecofeminist pedagogy argues that gender, race, class, and other forms of domination are deeply connected to oppression of the natural world, and that we should not elevate technology at the expense of ecology. This article also shares advice on how to incorporate critical sustainability frameworks, such as the Principles of Environmental Justice, into coursework across disciplines to help students reflect on the environmental impact of their consumption practices and habits. In all these ways, this commentary proposes an ecofeminist lens which can help instructors to appropriately ground new technologies like AI – which may seem abstract, disembodied, or virtual – within the physical and environmental present
System Characterization of a Distributed Pressure Sensor Array
No full textThis thesis contributes to ongoing research on bluff body aerodynamics at California Polytechnic State University, San Luis Obispo, by introducing a novel technique for the dynamic characterization of a commercially available digital MEMS pressure sensor. The experimental approach combines a new characterization method with comparisons to traditional techniques, offering a comprehensive evaluation of the sensor’s dynamic performance. Both non-parametric and parametric system identification methods are employed to assess the sensor’s frequency response. This thesis supports further investigations into unsteady aerodynamic forces and provides a cost-effective, hands-on platform for students to gain practical experience with instrumentation, system identification, and frequency-domain analysis
Deep Learning in Eye-Tracking Biomarkers for Anomaly Detection
No full textThis research project proposes the development of a novel, data-driven framework for detecting concussions using machine learning (ML) and deep learning (DL) models applied to high-resolution eye-tracking data. Unlike traditional concussion assessments that rely on subjective evaluations, this work seeks to identify objective, quantifiable biomarkers derived from ocular dynamics—such as saccadic velocity, smooth pursuit accuracy, and pupillary response—captured through advanced eye-tracking technology. The project will explore state-of-the-art feature extraction techniques and predictive modeling approaches to uncover subtle neuro-ocular signatures associated with mild traumatic brain injury. By combining principles from biomedical signal processing, artificial intelligence, and neurophysiology, this work advances current understanding of how concussions manifest in eye movement patterns and contributes to the interdisciplinary development of intelligent diagnostic systems. The outcomes have the potential to transform clinical and sideline concussion screening by enabling more accurate, timely, and scalable detection methods grounded in computational neuroscience
A Convolutional Neural Network Approach to Breast Cancer Tumor Boundary Detection
No full textBreast cancer is the second most common form of cancer and often goes undetected in its initial stages due to its subtle symptoms. As the tumors grow, they become more difficult to surgically remove with clean borders. To minimize the chance of recurrence, a 2D convolutional neural network is developed in this work for delineating tumor boundaries. Specifically, the U-shaped network model is trained, validated, and tested with longitudinal MRI scans of patients diagnosed with breast cancer and undergoing neoadjuvant therapy. For training, image masks were generated as ground truths using signal enhancement ratio segmentation, thresholding, and contour detection. The effectiveness of two lightweight models, with and without batch normalization and dropout layers, were compared for accuracy, loss, and processing times. The model including batch normalization and dropout layers outperformed the network without these functions; the enhanced model was additionally compared to several other more intricate machine learning algorithms for tumor-feature detection. The convolutional neural network designed for this work performed with higher accuracy and lower loss rates than the majority of the complex models, demonstrating that algorithm simplicity does not mean reduced functionality of the machine learning system
Motivation and Engagement in Outreach Activities
No full textIf the four elements of motivation and engagement are present in engineering outreach activities, does this affect the desire to become an engineer? Four elements: Autonomy, Purpose, Community, Mastery.
If the four elements of motivation and engagement are present/increased, and thereby motivation and engagement increase, do more students attending an engineering outreach event want to become engineers? If yes, then those activities where the four elements are increased/present should show an increase in students who are engaged and motivated to become engineers. If no, then the design of activities is not a major deciding factor in whether a student wants to become an engineer after the activity
Building a Smart Transportation Network to Prevent Multi-Vehicle Collisions During Sudden Slowdowns
No full textThis proposed SURP project aims to design and evaluate a smart transportation network capable of preventing multiple-vehicle collisions due to sudden slowdowns in traffic. This project will simulate abrupt braking scenarios and implement adaptive vehicle-to-vehicle (V2V) communication protocols. By enhancing real-time awareness and responsiveness among vehicles, the system will reduce pileup risks and improve road safety