55453 research outputs found
Sort by
MACHINE LEARNING MODELS LEVERAGING PATIENT-SIMILARITY AND CLINICAL TEMPORALITY FOR DISEASE PROGNOSES
Get PDFElectronic Health Records (EHRs) constitute a comprehensive and high-dimensional repository of clinical data, encompassing a wide array of patient-level information such as diagnoses, procedures, medications, laboratory results, and unstructured clinical narratives. These data hold immense potential for advancing predictive modeling in healthcare, including tasks such as disease progression modeling, hospital readmission prediction, and length of stay (LoS) estimation. However, the intrinsic complexity of EHR data—manifested in its heterogeneity, sparsity, and temporal dynamics—poses significant analytical challenges that limit the generalizability and interpretability of conventional machine learning models. Recent methodological advancements in deep learning and graph-based learning, particularly Graph Neural Networks (GNNs), have opened new avenues for representing and analyzing the multifaceted structure of EHR data. In this dissertation, we propose a set of enhancements to the Patient Similarity Graphs (PSGs) framework, a graph-based paradigm for modeling patient-patient relationships by integrating heterogeneous clinical information. Our contributions span multiple dimensions of representation learning and predictive modeling, targeting the improvement of clinical outcome predictions across various healthcare applications. We begin by benchmarking traditional machine learning models on tabular EHR representations, followed by the development of homogeneous PSGs and their use in GNN-based models for LoS prediction. Subsequently, we introduce a PageRank-based similarity framework for diagnosis prediction in both static and temporal contexts, leveraging sequential visit data. Furthermore, we conduct a comprehensive investigation into link prediction methodologies within biological networks, analyzing their effectiveness in relation to network assortativity. Finally, we evaluate multiple data fusion strategies to construct both patient-level and visit-level similarity graphs for enhanced temporal and static prediction tasks. By integrating structured and unstructured features into unified similarity graphs, and by exploring the temporal evolution of patient trajectories, our work demonstrates the efficacy of graph-based approaches in modeling complex EHR datasets. The proposed enhancements to the PSG framework not only improve predictive performance across a range of clinical tasks but also offer a scalable and interpretable methodology for precision medicine and personalized healthcare analytics
Exploring Collaboration In Interdisciplinary EI/ECSE Programs: A Qualitative Study On Service Provider Preparation
Get PDFInterdisciplinary preparation is critical for early childhood special education professionals, yet few programs are designed to support such collaboration across disciplines. This study aimed to explore how interdisciplinary EI/ECSE preparation programs support collaboration. The study was guided by the following research questions: (1) How are program components designed to support the development of collaboration skills? (2) What is the student experience of collaboration in the program? (3) What connections exist between program components, student experiences, and the development of collaboration skills? Using a qualitative multiple-case study approach grounded in ecological systems theory, data were collected from program documents and interviews with faculty and students from two preparation programs. Findings suggest that, through shared classes, interdisciplinary seminars, fieldwork, and supportive relationships, interdisciplinary EI/ECSE preparation programs contribute to the development of collaboration skills. Implications include suggestions for planning and implementing future programs, as well as recommendations for future research that investigates outcomes while considering factors such as disciplinary background. This study advanced knowledge on interdisciplinary EI/ECSE programs by linking students\u27 development of collaboration skills to program experiences and by generating an expanded conceptual framework that illustrates connections between skills, program components, disciplinary differences, and data across multiple levels
An Asthma Patient Education Pilot: Utilizing a Hand-Held Tablet to Enhance Self-Management Knowledge
Get PDFPurpose:
This study aimed to evaluate the impact of pre-recorded educational videos on a hand-held tablet for improving asthma self-management knowledge. Content focused on disease processes, medication management, and medical device usage in an adult asthma clinic at an academic medical center.
Methods:
Pre-recorded asthma education videos were created by study personnel and uploaded to the institutional research electronic data capture (REDCap) platform. Adult participants (≥18 years, English-speaking, with a new or existing asthma diagnosis) were enrolled at an outpatient asthma clinic. The Asthma Self-Management Questionnaire, a validated and reliable tool, was administered before and after participants watched five 1-minute educational videos (total duration≈ 5 minutes) on topics such as asthma pathophysiology, triggers, flare-ups, inhaler use, asthma action plans, and peak flow monitoring. After informed consent, participants watched the videos on a hand-held tablet and completed the questionnaire again. A follow-up additional survey assessed tablet usability and participant satisfaction. Knowledge scores were calculated based on the percentage of correct answers.
Results:
Thirty-seven participants were enrolled in the study. The participants were predominantly female (78.4%), with most participants identifying as Black or African American (54.1%) or White (40.5%). The average age of participants was 47.8 years (SD = 18.8), with a median of 52 years. Following video viewing, the mean knowledge score significantly increased to 69.6 (SD = 16.9), with a mean improvement of 15.0 (SD = 15.1) points. A paired t-test revealed a statistically significant improvement (p \u3c 0.05), indicating that the educational videos effectively enhanced asthma self-management knowledge. Usability and satisfaction scores were high across all participants, with both younger and older individuals reporting similar evaluations of the device’s performance and educational value.
Conclusions:
This pilot study demonstrates that watching pre-recorded educational videos on a hand-held tablet significantly improves asthma self-management knowledge. The device was well-received, with high satisfaction and usability scores among a diverse group of participants.
Clinical Implications:
As clinicians increasingly manage complex, chronic diseases within a time-constrained and resource limited environment, providing comprehensive patient education becomes a challenge. However, enhancing patient education and health literacy is essential for improving outcomes in chronic disease management. This study demonstrates that pre-recorded video education, focused on disease-specific topics, is a feasible and effective way to augment outpatient visits. By supplementing in-person interactions, this method not only improves patient understanding but also enhances health literacy, ultimately supporting enhanced self-management and improved outcomes in chronic disease care
A Novel Perspective: Utilizing Mixed Methods Approach to Identifying Critical Intervention Periods and Factors that Impact Quality of Life for Pediatric Inflammatory Bowel Disease
No full textPediatric inflammatory bowel disease (IBD) affects not only physical health but also emotional and social well-being, often leading to a reduced quality of life. Despite medical advances, over half of patients eventually lose response to medication, and many—especially children—receive little support beyond prescriptions. Drawing from her own experience as a pediatric IBD patient, Caterina LaRocca’s research seeks to identify critical intervention periods and factors that improve quality of life for children with IBD. Using a mixed-methods approach, her first two qualitative studies will capture patient-informed perspectives through interviews with children and adults diagnosed in childhood, focusing on unmet needs in mental health, social support, and daily functioning. A subsequent quantitative study will assess modifiable factors that influence transition readiness from pediatric to adult care, such as self-efficacy and disease knowledge. Together, these studies aim to build an evidence base for more holistic, patient-centered interventions—ensuring that children with IBD are supported as whole individuals, not defined solely by their disease
Orienting New Researchers Across Multiple Pathways: Bridging Research Resource Gaps for New Faculty at VCU
Get PDFNew faculty researchers at VCU, who possess diverse research interests or requirements, often find it challenging to navigate the VCU environment in search of research resources necessary to initiate their work. The majority of available research resources highlighted on central VCU pages primarily cater to biomedical research, inadvertently marginalizing those in less prominent fields. Resources in other departments, such as fine arts and education, are limited and often restricted to the departmental or school level. This raises the essential query: How and where should I begin my research? The project employs a faculty and staff survey aimed at assessing the existing research training during orientation and providing recommendations for future steps
VCU SciCom: A Federated Unit for Community-Engaged Science Communication
Get PDFConfidence and trust in science have declined in recent years, in part due a lack of understanding of how science works and that knowledge is not immutable but constantly evolving. At the same time the rise in social media has led to the spread of misinformation and changes in the media landscape have made it difficult to disseminate the correct information to channels that reach the greatest numbers. While most Americans see scientists as intelligent, less than half (45%) say that they are good communicators. As a public university, Virginia Commonwealth University (VCU) serves the public good by ensuring that it remains a trustworthy source of medical and scientific information and that the science we do is seen as relevant to the communities we serve. VCU employs renowned scientists generating world-class research and while findings are disseminated through press releases and other channels, these findings are not always accessible or understandable to all. This project seeks to close that gap by proposing a federated science communication unit that translates the science we do into plain-language, culturally relevant, and community-relevant outputs, while doing so in a way that meets people where they are. Over a two-year initial phase, the project recommends that VCU establish a steering committee, develop standardized plain-language research summary templates, train faculty and trainees, and recruit trusted community ambassadors to disseminate findings. Recommended deliverables include a centralized online platform of research summaries, community listening sessions, tailored outreach campaigns, and annual forums to promote dialogue between researchers and the public. Proposed outcomes will be measured through community surveys, equity analyses, and engagement metrics. By Year 2, the project will evolve into a fully functioning Federated Science Communication unit housed under the Office of the Vice President for Research and Innovation (OVPRI), with long term sustainability supported by the institution. This initiative will strengthen community trust, expand research participation, and elevate VCU’s reputation as a national leader as a community facing public institution serving the public good
Junior Recital, Wes Castle, video
Get PDFJunior Recital VideoWes Castle, guitar, videowith Malik Gray, tenor saxophone; Kevin Granados, tenor saxophone; Hepburn Hull, piano; Libby Mullins, vocals; Avery Waddell, bass; and April Freeman, drumsTuesday, September 2, 2025 at 6:30 p.m.Recital HallJames W. Black Music Center1015 Grove Avenue | Richmond, Virgini
Oral Health Impact and Cost-Effectiveness of Resealing First Permanent Molars
Get PDFIntroduction: This study examined the cost-effectiveness and health impact of various dental sealant resealing scenarios for first permanent molars from the payer perspective.
Methods: The authors used a discrete event simulation model for 4,000 recently sealed first molars in a hypothetical cohort of 1,000 children aged 7 years over a 10-year time horizon. Scenarios included a base case of no resealing, resealing at first dental visit, and resealing at the age of 12 years.
Results: Without resealing, only 34.88% of teeth retained sealants after 10 years, whereas resealing at the age of 12 years increased this to 53.05%, and resealing at first dental visit increased it to 43.48%. Resealing at age 12 years significantly reduced decay experience to 26.65%, and resealing at first dental visit reduced it to 25.30%, compared with 41.15% for no resealing. The number of accrued disability-adjusted life years was highest for no resealing (5.35). This number reduced to 3.52 for resealing at age of 12 years and 3.19 for resealing at the first dental visit. Cost-effectiveness analysis showed that the most effective resealing scenario, resealing at the first dental visit, also had the lowest cost per averted cavity (6,753).
Conclusions: This simulation model indicates that resealing significantly reduces dental decay over 10 years at reasonable costs, using conservative modeling assumptions. These findings suggest that insurance companies and Medicaid could benefit from implementing resealing policies to improve dental health outcomes by reducing caries and filling costs
