Sinclair Community College

We introduce PASTA (Perceptual Assessment System for explanaTion of Artificial Intelligence), a novel human-centric framework for evaluating eXplainable AI (XAI) techniques in computer vision. Our first contribution is the creation of the PASTA-dataset, the first large-scale benchmark that spans a diverse set of models and both saliency-based and concept-based explanation methods. This dataset enables robust, comparative analysis of XAI techniques based on human judgment. Our second contribution is an automated, data-driven benchmark that predicts human preferences using the PASTA-dataset. This scoring called PASTA-score method offers scalable, reliable, and consistent evaluation aligned with human perception. Additionally, our benchmark allows for comparisons between explanations across different modalities, an aspect previously unaddressed. We then propose to apply our scoring method to probe the interpretability of existing models and to build more human interpretable XAI methods.

<title>Abstract</title> Background Type 2 Diabetes Mellitus (T2DM) is a priority health challenge due to its prevalence and the clinical impact of poor glycemic control. Behavioral determinants play a central role in modulating metabolic outcomes within nursing practice. Aim: To determine the impact of behavioral determinants on glycemic control in adults with T2DM from a translational nursing perspective. Methods: Systematic review following PRISMA 2020 guidelines. Observational studies and randomized clinical trials evaluating self-care, self-efficacy, health literacy, and glycemic control (HbA1c) were included. Data were integrated using traceability and critical evidence matrices to normalize heterogeneous statistical metrics. Results: Nursing-led interventions and the strengthening of self-efficacy and self-care are associated with clinically significant improvements in HbA1c. Conversely, low health literacy was identified as a critical risk factor for metabolic poor control. Conclusion: Behavioral determinants are the main modulators of glycemic control in T2DM. Translational nursing emerges as a strategic axis for transforming scientific evidence into high-impact clinical interventions.

Abstract Every year in the US, 30% of older adults experience falls, resulting in three million visits to emergency departments. Fall-monitoring technologies have proven effective in improving the quality of life for older adults, allowing them to age in place. Though these technologies enhance safety and keep caregivers informed, they also raise concerns regarding autonomy and privacy. However, there is limited understanding of the perspectives of older adults living with these technologies. We used an interpretive phenomenological approach, interviewing eight older adults, with and without dementia, residing in two independent living facilities who had experienced fall-detection technologies for the previous 9-12 months. The interviews focused on understanding how technology influences their self-identity, ability to live independently, safety, and age-in-place. To ensure trustworthiness, field-notations, reflexive journaling, and participant check-backs were used. Several themes emerged reflecting the influence of technologies on lived experiences: the freedom to age in place, the necessity for active intervention, individual approaches to technology based on personality/cognition, and feelings of discomfort and fear. Older adults were more inclined to engage with monitoring technologies if they were open to intervention, had their concerns addressed, and perceived the technology as useful for enabling aging-in-place. Older adults, with and without dementia, are often the target audience for monitoring technologies, yet their perspectives are frequently overlooked. This study significantly enhances the understanding of effective implementation strategies for monitoring technologies. It serves as a valuable resource for individuals and facilities seeking successful integration of fall-detection and prevention technologies to support older adults to age-in-place.

This volume showcases the research, teaching, and community engagement of fellows from the Mellon/ACLS Community College Faculty Fellowship program, which from 2018 to 2025 supported 110 faculty across 61 institutions. Edited by Carmen Carrasquillo (San Diego Miramar College) and Brian Stipelman (Frederick Community College), the collection highlights the challenges and opportunities faced by two-year-college scholars, documenting how they integrate research into the classroom and serve diverse student communities. More than a record of individual projects, the publication celebrates the vital role of community college faculty as teacher-scholars whose contributions enrich the humanities and higher education at large.

This survey presents a comprehensive analysis of ten cutting-edge multimodal datasets that address critical challenges in domains such as structural biology, healthcare, computer vision, audio-visual understanding, and robotics. By examining their design objectives, data properties, and evaluation scopes, we identify emerging trends in fairness-aware learning, compositional generalization, and cross-modal reasoning. Our findings highlight how thoughtful dataset construction shapes the development of robust, interpretable, and human-aligned AI systems. Benchmark design is therefore a central driver of trustworthy machine intelligence, particularly in aerospace applications where such systems are critical for autonomous operation and mission-critical decision-making.

Abstract This project's goal is to improve critical thinking in undergraduate engineering students through a new educational intervention aimed at enhancing ethics and professional responsibility. Developing an understanding of how engineering students perceive the broader impacts of their chosen profession, especially as articulated through the lens of ethics and professional responsibility, is an important first step in addressing professional formation. The current model of engineering education, focused primarily on technical proficiency, leaves little room for the structured and integrated exploration of these broader impacts. The intervention that was evaluated in this study involved student-group discussions that were centered around professionally produced narratives in an audio format. Students were required to listen to the narrative, respond to focus questions, engage with their peers' responses to the questions and then reflect on the experience. After completing this assignment for three narratives, students (N=45) participated in a separate discussion involving the ethical issues and broader impacts of engineering work that they encountered while working on their senior design projects. The same exercise was exercise was completed by a comparison group (N=49) that was not exposed to the intervention. Our results indicate that exposure to the critical narrative intervention did enhance students' abilities to identify ethical issues and broader impacts of engineering work (p<.05).

Abstract Sixty-eight years after the Brown v. Board of Education decision led to a national reduction of teachers of color, even as numbers of students of color grew, a collaborative effort to racially re-diversify the education profession is forming in the Dayton, Ohio region. This article describes this ongoing movement to increase the number of racially and ethnically marginalized educators, led by seven school districts, five institutions of higher education (IHEs), a county educational services center, and two nonprofit organizations. Initial work was supported by leveraging grant funds. The project seeks to (1) identify seventh- through eleventh-grade students interested in becoming educators, (2) direct students into model pathways and critical supports from high school through college graduation and licensure, (3) offer support for minoritized future and new teachers that fosters confidence and promotes belonging, and (4) facilitate the emergence of collectively-designed professional development for continuous renewal of culturally responsive and inclusive cultures in all education spaces. Strategies include identifying prospective teachers within high schools, establishing peer cohorts, facilitating critical mentoring, mapping grow-your-own pipelines and pathways, providing support at crucial stages in college, developing positive cultures in programs and districts, facilitating ongoing professional development that centers marginalized perspectives, and providing logistical support for emerging networks and organic affinity groups. Lack of support, student debt, isolation, discrimination, and unwelcoming and unresponsive school environments are common barriers that require ongoing dismantling.

This book examines the detailed process to develop fully autonomous drones that incorporates object-oriented design and cognitive deep learning systems.