VTechWorks
VTechWorks provides global access to Virginia Tech scholarship, including journal articles, books, theses, dissertations, conference papers, slide presentations, technical reports, working papers, administrative documents, videos, images, and more by faculty, students, and staff. Faculty can deposit items to VTechWorks from Elements, including journal articles covered by the University open access policy. Email vtechworks@vt.edu for help.
Open Access Policy
Virginia Tech's open access policy enables researchers to deposit the accepted version of scholarly articles with no embargo.
Theses and Dissertations
Virginia Tech was first in the world to require ETDs in 1997, and continues to add scans of older theses and dissertations.
Open Textbooks
More than 50 freely available and openly licensed textbooks are among our most downloaded items.
Communities in VTechWorks
Select a community to browse its collections.
Recent Submissions
A Demand-Controlled Ventilation (DCV) Systems Approach Towards Energy-Efficient Management for Improved Indoor Air Quality (IAQ)
GhazanfariMotlagh, Niloufar (Virginia Tech, 2026-09-23)
Indoor Air Quality (IAQ) is a critical concern in commercial buildings and specifically for educational environments, as poor ventilation and inadequate air filtration have been linked to adverse health effects and reduced cognitive performance. In commercial spaces, insufficient IAQ can lead to occupants' discomfort, increased sick leave, and lower productivity, whereas in educational environments, students experience impaired concentration, decreased learning outcomes, and higher absenteeism due to pollutants such as CO₂, volatile organic compounds (VOCs), and particulate matter. Studies indicate that elevated CO₂ levels and poor ventilation in classrooms can negatively affect students' cognitive functions, including decision-making and problem-solving skills.
Despite growing awareness of IAQ's impact, ventilation strategies have not kept pace with advancements in building technology. Smart buildings, which integrate automated HVAC systems, smart windows, photovoltaic panels (PVs), and geothermal solutions, are designed to enhance energy efficiency while maintaining occupant comfort. Among heating, cooling, and ventilation strategies, the most significant advancements have been in heating and cooling systems, whereas ventilation improvements have lagged behind, This may be due to the fact that the precise control of heating and cooling systems was seen as low hanging fruit to achieve higher savings. Since achieving adequate ventilation required to maintain the indoor air quality is much less understood and researched, current practice still follows prescriptive rules. An effective ventilation system should be able to provide a reasonable balance between pre-conditioned fresh air supply rates and energy usage.
Conventional ventilation systems with manual ON/OFF control have the limitation that the total ventilation rate cannot fluctuate by demand since it only has two operation modes. A simple ON/OFF control (fixed volume ventilation) approach always needs to pick a relatively high ventilation rate at the expense of consuming more energy to ensure an adequate degree of air quality is provided during peak times. Moreover, conventional ventilation systems are incapable of giving control over specific settings or altering fresh air supply in an office building depending on the dynamic occupancy of particular rooms. This mismatch between ventilation supply and real-time demand often leads to significant energy waste and undermines efforts to improve building energy efficiency.
In recent years, Demand-Controlled Ventilation (DCV) systems have emerged as a promising solution to address these inefficiencies. DCV systems utilize real-time data from indoor air quality sensors to dynamically adjust the ventilation rates based on occupancy usage and indoor environmental conditions. This approach enables a more precise balance between maintaining indoor air quality and minimizing energy consumption. However, the effectiveness of DCV systems relies heavily on the accuracy of CO₂ or occupancy prediction models and the responsiveness of ventilation controls, which present ongoing challenges in system design and implementation.
Machine learning (ML) techniques have shown great potential in addressing the limitations of traditional physics-based CO₂ prediction models. By analyzing historical and real-time environmental data, ML models can learn complex patterns and relationships between variables such as occupancy, temperature, humidity, and ventilation operations. These advanced models can significantly improve the accuracy of CO₂ predictions, enabling more effective control strategies for ventilation systems. Furthermore, integrating ML-based predictions with DCV systems allows for real-time optimization of ventilation rates, ensuring that both energy efficiency and indoor air quality objectives are met.
Despite these advancements, the implementation of DCV systems in existing buildings often faces challenges, particularly in retrofit scenarios. Retrofitting an existing HVAC system to incorporate DCV requires careful consideration of factors such as ductwork modifications, sensor placement, and system calibration. Additionally, the lack of standardized guidelines for DCV retrofits can make the decision-making process complex for building professionals.
Therefore, there is a need for a control strategy that enables us to anticipate and identify fluctuations in the trends of CO₂ levels. This approach will enable the modification of ventilation rates in real-time while concurrently reducing energy consumption, thus providing an optimized solution for indoor air quality, energy efficiency, and occupant well-being. Studies indicate that maintaining low CO₂ levels through optimized ventilation significantly enhances decision-making, focus, and overall cognitive performance. By leveraging machine learning algorithms alongside advanced sensor technologies, next-generation DCV systems can transform building ventilation practices, fostering smarter, healthier, and more sustainable indoor environments.
The Impact of Tamoxifen on Cardiac Electrophysiology in the Context of the Tamoxifen-Inducible Cre-Loxp System in Mice
Albrecht, Maxwell Reid (Virginia Tech, 2026-09-22)
Sudden cardiac death is one of the leading causes of death worldwide, with lethal ventricular arrhythmias being likely responsible for most incidents. The frequency of sudden cardiac death is increasing, presenting the need for more research into the causes and treatments of these deadly arrhythmias. As arrhythmia research increases to meet the need for more effective treatments, there is a distinct need for increased scrutiny into the possible confounding variables present in current models of arrhythmogenic disease.
One of the most common and effective models for arrhythmia research is mice, utilizing the tamoxifen-inducible cardiac specific Cre-loxp system. In this system, tamoxifen is used to activate the Cre recombinase protein, made to express only in the heart. This protein targets and removes two inserted artificial loxp sites and the sequence between them, resulting in a gene knockout or other manipulation. This system can be used to target specific proteins known to contribute to arrhythmia formation, while avoiding developmental complications common in constitutive knockouts of cardiac proteins.
The cardiac-specific Cre-loxp system is known to have off-target effects, including inflammation, fibrosis, and reduced cardiac performance. Tamoxifen also has effects on cardiac electrophysiology, with the ability to block sodium channels, which can slow conduction and promote arrhythmias. As many studies focus on tamoxifen in the context of its use in human patients, it is unknown whether tamoxifen on its own, as used in the mouse Cre-loxp system, can create an arrhythmogenic phenotype. We conducted studies to determine how tamoxifen impacts cardiac electrophysiology and the expression of key proteins responsible for conduction in an ex vivio mouse model. We found that tamoxifen could slow conduction, and change action potential duration characteristics, particularly at shorter paced cycle lengths. This indicates that tamoxifen can interfere with cardiac electrophysiology, and that these effects need to be controlled for in arrhythmia research using models which are exposed to tamoxifen.
Scaling Up Complex Sensemaking and Problem-Solving Through Crowdsourcing and Human-AI Collaboration
Mukhopadhyay, Anirban (Virginia Tech, 2026-09-22)
Open Source Intelligence (OSINT) is widely used to collect and analyze publicly available information for applications such as countering disinformation, cybersecurity vulnerability assessment, journalism, law enforcement, and human rights investigations. However, scaling OSINT investigations remains difficult because analysts must process large volumes of unstructured and ephemeral data, verify uncertain information, operate diverse technical tools, and coordinate work across teams. These challenges are compounded by gaps in technical expertise and by the risks of introducing unreliable automation.
In this dissertation, I investigate how such collaborative sensemaking and problem-solving tasks can be scaled through two complementary approaches: crowdsourcing to augment expert investigations and human--AI collaboration to support individuals and teams. First, I developed OSINT Research Studios, a flexible crowdsourcing framework in which trained novices support professional investigators through structured discovery and verification tasks under expert supervision. Second, I introduced the OSINT Clinic and conducted a longitudinal co-design study to identify the challenges novices face during collaborative vulnerability assessments and examine how generative AI can support planning, data collection, processing, analysis, dissemination, collaboration, and leadership. Third, I studied proactive generative AI agents in time-sensitive, co-located problem-solving by comparing facilitator and peer roles and measuring their effects on group performance, workload, communication, and coordination. Finally, I developed the OSINT Research Platform, a transparent and context-aware tool-calling LLM agent that can orchestrate OSINT tools, preserve human oversight, and provide next-step suggestions based on shared team activity.
Overall, these studies contribute sociotechnical frameworks, empirical findings, and design requirements for scaling collaborative investigations. I show that effective scaling requires more than increasing participation or automating tasks; it requires systems that preserve context, verification, transparency, ethical boundaries, and human judgment. The dissertation offers design recommendations trustworthy human--AI collaboration with evolving roles of AI in collaborative work---from tool and personal assistant, to proactive teammate, to agentic collaborator.
Inviting Students to the Table: Negotiating Power in Course Design
Scherer, Hannah H. (College STAR Faculty Development Modules & Case Studies, 2021-07)
From Biofilms to Bubbles: Dispersion of Bacteria from Biofilms Using Histotripsy
Hoffman, Carson Lee (Virginia Tech, 2026-09-21)
Orthopedic infections are difficult to treat and diagnose because bacteria often persist within biofilms, fibrin networks, host tissue, and other protective infection-associated matrices. These structures limit antimicrobial penetration, reduce bacterial susceptibility, and make viable bacteria difficult to recover for culture. This thesis investigated whether focused ultrasound technologies could mechanically disrupt these protective bacterial environments to improve both antibiotic-mediated bacterial killing and culture-based diagnostic recovery.
The first objective was to evaluate histotripsy as an adjunctive strategy to improve antibiotic killing of biofilm-associated bacteria in equine synovial environments. Staphylococcus aureus ATCC™ 25923 and an agrC knockout strain were grown as biofilm-associated aggregates in either biofilm media or pooled equine synovial fluid and treated with histotripsy, proteinase K (200 µg/ml), amikacin sulfate, or combination treatments. Histotripsy consistently disrupted biofilm structure and produced greater dispersal than untreated controls and proteinase K (200 µg/ml), as measured by optical density. Amikacin alone had limited activity against protected biofilm-associated bacteria, while combination treatment with histotripsy and amikacin produced greater bacterial killing than amikacin alone. In biofilm media, combination treatment reduced viable bacterial counts by up to approximately 6 log₁₀ CFU/mL compared with untreated controls. Bacterial killing was less pronounced in synovial fluid, where viable bacteria persisted despite dispersal and antibiotic exposure. These findings suggest that mechanical disruption of biofilm-associated aggregates can improve antibiotic access and enhance bacterial killing, while also demonstrating the protective effect of synovial fluid and infection-associated matrices.
The second objective was to evaluate Focused Ultrasound Extraction (FUSE), as a rapid dispersal method to improve bacterial recovery from biofilms and infected tissue. Biofilms of Staphylococcus aureus ATCC™ 25923 and Pseudomonas aeruginosa PAO1 were grown for 24 hours in RPMI containing 20% equine plasma and treated with proteinase K (200 µg/ml), sonication, or FUSE at increasing total pulse doses. For S. aureus ATCC™ 25923, FUSE at 40 cycles, 200 Hz pulse repetition frequency, and 20,000 total pulses produced the greatest mean recovery of culturable bacteria, while higher pulse doses reduced recoverability. For P. aeruginosa PAO1, lower FUSE doses did not significantly increase recovery, and higher pulse doses significantly reduced recoverability. These findings demonstrate that FUSE dose must be optimized to favor bacterial release while preserving viability, and that the effective dose window differs by organism.
FUSE was then evaluated using equine-derived clinical isolates of Escherichia coli, Klebsiella pneumoniae, and Staphylococcus aureus. FUSE significantly improved recovery from K. pneumoniae M26-1142 at 10,000 pulses and from S. aureus M22-2451 at 20,000 pulses, while no significant treatment effect was detected for E. coli M25-0692 under the conditions tested. These isolate-dependent differences likely reflect differences in biofilm structure, density, and matrix composition. S. aureus formed a visibly more robust and structured biofilm than K. pneumoniae or E. coli, creating a greater opportunity for dispersal-based methods to improve bacterial recovery. In contrast, relatively high bacterial recovery from untreated controls for some Gram-negative isolates suggests that these organisms may have been more readily recoverable without active dispersal in this model.
Finally, FUSE was evaluated using human periprosthetic joint infection tissue samples and compared with an enzymatic dispersal agent. FUSE rapidly disrupted infected tissue and released culturable bacteria within minutes. In the majority of samples, FUSE produced greater colony recovery per gram of tissue than enzymatic dispersal, while requiring only a brief treatment period. These findings support the potential use of FUSE as a rapid culture-preparation strategy for infected tissue.
Together, these studies support focused ultrasound as a versatile mechanical disruption platform for orthopedic infections. In the therapeutic setting, histotripsy improved antibiotic-mediated bacterial killing by disrupting protected bacterial aggregates. In the diagnostic setting, FUSE improved recovery of viable bacteria from biofilms and infected tissue when delivered within an organism-appropriate dose range. These findings address a shared clinical problem: bacteria embedded within protective matrices are difficult to kill and difficult to culture. Further optimization and validation using larger clinical sample sets may support future development of focused ultrasound strategies for both treatment and diagnosis of biofilm-associated orthopedic infections.