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 40 freely available and openly licensed textbooks are among our most downloaded items.
Communities in VTechWorks
Select a community to browse its collections.
Recent Submissions
Developing a Local Networking Group for Adults with Celiac Disease in Southern Virginia
Standbrook, Abigail (Virginia Tech, 2025-06-26)
Celiac disease is an autoimmune disease affecting the microvilli of the small intestine.
Celiac disease requires a strict diet free of wheat, rye, and barley. Even a minimum exposure of a
few crumbs can cause microvilli damage and can cause symptoms for up to two weeks.
Individuals with celiac disease often report decreased quality of life due to the social and
emotional challenges of maintaining a strict gluten-free diet. Support groups have been shown to
be an effective way for individuals to navigate social and emotional challenges for disease
management. The purpose of this project was to develop a support group for individuals with
celiac disease in southern Virginia to address emotional support, access to evidence-based
research, and build empowerment to adhere to a strict gluten-free diet. Support groups
The pilot program was delivered as a virtual support group. It focused on providing
emotional support and empowerment by discussing hidden sources of gluten, balancing a healthy
diet, and social navigation. Additionally, it served as a safe space for participants to feel heard,
share their challenges, and learn from others.
Evaluation surveys indicated that participants felt heard and their learning needs were
met by the support group. Implications arose as only one individual participated. The results are
limited but support the existing literature that virtual support groups can provide emotional
support for adults with celiac disease.
Investigation of Future Voluntary Movement Prediction for Pathological Tremor-Alleviating Exoskeletons
Ding, Zijian (Virginia Tech, 2025-07-01)
Pathological tremor, a common neurological disorder, significantly impacts patients' daily quality of life and causes difficulties with performing simple daily tasks. Those tremors usually interfere with the patient's fine motor control and may also cause psychological anxiety and social barriers. Traditional treatments, such as medication and physical therapy, can alleviate symptoms to a certain extent; however, their effects are limited, and they may also have side effects. Therefore, rehabilitation exoskeletons have emerged as an assistive technology and have become an essential supplement to traditional treatments. Then, performance optimization is particularly critical to fully realizing the potential of exoskeletons. The ideal tremor suppressor exoskeleton not only needs to suppress tremors effectively but also must be able to distinguish and predict the patient's autonomous movements. These requirements can ensure that the exoskeleton minimizes the influence of involuntary tremors while facilitating the patient's voluntary movements, enabling patients to experience a smooth and natural operational experience similar to that of normal movement.
The wrist is a critical component of human operational capability, with its flexibility and precision playing an important role in daily activities. However, it is also a common site for pathological tremors. Our research laboratory has developed a wearable exoskeleton termed TAWE to address this issue. TAWE uses a 6-degree-of-freedom (DOF) rigid link mechanism, which can precisely replicate the natural range of motion of the wrist while simultaneously providing real-time suppression of pathological tremors without compromising the user's freedom of movement. Therefore, we developed a deep learning model based on a convolutional neural network (CNN) and self-attention mechanism to accurately extract and predict patients' voluntary movement intentions from tremor-affected motion data. This model enables real-time motion planning for the exoskeleton, achieving both tremor suppression and zero-latency performance. This model is capable of directly predicting voluntary movement trajectories approximately 100 milliseconds in advance from real-time input data.
Finally, we comprehensively evaluated the model's performance and its real-time capabilities when integrated into the exoskeleton system through simulation experiments. Overall, the CNN-Self-Attention-based model has strong performance and can predict autonomous motion trajectories for the next 100 milliseconds in real-time, regardless of whether the input data included tremor interference. However, the results also revealed certain model limitations under extreme conditions, such as high-frequency and large-amplitude tremors. In these cases, the output trajectory remained insufficiently smooth even after processing, resulting in slight stuttering during exoskeleton movement. These problems need further research and improvement.
Architectural Amalgam
Chu, Yu (Virginia Tech, 2025-07-01)
Building is the container of people, furniture, pets and plants. But also the container of stories, time, and memories. To recall memories, architecture offers the possibility of contrast of old buildings and new parts to unfold and highlight what exists in a distinctive way, to make the memories more clear.
This thesis uses the notion of amalgam as an example, where new material, natural light, and special spaces form a contrast, to evoke our memories.
Efficient Reinforcement Learning for Control
Baddam, Vasanth Reddy (Virginia Tech, 2025-07-01)
The landscape of control systems has evolved rapidly with the emergence of Reinforcement Learning (RL), offering promising solutions to a wide range of dynamic decision-making problems. However, the application of RL to real-world control systems is often hindered by computational inefficiencies, scalability issues, and a lack of structure in learning mech- anisms. This thesis explores a central question: How can we design reinforcement learning algorithms that are not only effective but also computationally effi- cient and scalable for control systems of increasing complexity? To address this, we present a progression of approaches—starting with time-scale decomposition in small- scale systems and moving towards structured and adaptive learning strategies for large-scale, multi-agent control problems. Each chapter builds upon the previous one by introducing new methods tailored to the complexity and scale of the environment, culminating in a unified framework for efficient RL-driven control
Toward Robust and Generalizable Spatiotemporal Modeling for Tasks beyond Forecasting and Classification
Sun, Yanshen (Virginia Tech, 2025-07-01)
In spatiotemporal data mining, building models that are robust and generalizable across complex, non-ideal conditions is crucial for real-world deployment. While many existing methods perform well on benchmark datasets, they often assume clean, stationary, and uniformly sampled data, limiting their effectiveness in practical scenarios. In diverse operational domains such as traffic systems, neurophysiological monitoring, and drilling operations, spatiotemporal data is often noisy, irregular, and subject to distribution shifts — exposing the brittleness of conventional forecasting and classification pipelines.
This dissertation advances spatiotemporal modeling by addressing three critical challenges that frequently arise in real-world applications but fall outside the scope of traditional forecasting and classification: anomaly detection, domain adaptation, and causal discovery. It systematically examines these issues across three cross-disciplinary application domains and proposes targeted, scenario-specific solutions: textbf{(1) Anomaly Detection:} We develop spatiotemporal anomaly detection frameworks to identify and mitigate irregularities in both traffic forecasting and EEG signal classification tasks. textbf{(2) Domain Adaptation:} We design and evaluate domain adaptation strategies that enable robust cross-patient EEG classification, addressing inter-subject variability and enhancing generalization across diverse patient data. textbf{(3) Causal Discovery:} We integrate causal discovery techniques into drilling fluid loss prediction workflows to uncover latent causal relationships, thereby enhancing the extrapolation capabilities of fine-tuned time series foundation models in previously unseen scenarios.
Anomaly detection techniques are applied across three distinct tasks: detecting abnormal traffic sensor measurements, predicting the impact of traffic incidents, and classifying EEG signals for depression diagnosis. In the context of traffic sensor anomaly detection, key challenges include (1) modeling spatiotemporal dependencies to capture irregular patterns, (2) distinguishing implicit anomalies from regular fluctuations, and (3) maintaining robustness in the absence of reliable reference data. To address these issues, we propose S-DKFN, an unsupervised model that fuses spatial and temporal features to uncover complex anomaly patterns. It incorporates dilated temporal convolutional networks (TCNs), an encoder-decoder structure for multiscale representation learning, and leverages Kalman filtering principles for model fusion to improve robustness and accuracy.
Traffic incident impact (TII) prediction also presents significant modeling challenges, particularly due to the dynamic nature of real-world traffic networks. Prior studies have often (1) overlooked systematic quantification of spatiotemporal TII and suffered from a lack of open benchmark datasets, (2) struggled to adapt attention mechanisms to capture interactions over time-varying road networks, and (3) failed to identify task-relevant substructures in space and time. To overcome these limitations, we first provide a formal quantification of TII and release two curated open-source datasets. We then propose two novel models: the RAS-Transformer, designed to locate affected sub-graphs, and the IST-Transformer, which leverages importance-score-based adversarial training to focus attention on sensors most impacted by incidents.
EEG signal classification for depression diagnosis poses its own set of difficulties. Existing methods typically (1) struggle to extract meaningful patterns from noisy and non-stationary EEG signals, (2) rely heavily on manual preprocessing and handcrafted features, and (3) fall short in capturing the spatial and temporal dependencies intrinsic to neural activity. In response, we propose a novel spatiotemporal deep learning model tailored for depression-related EEG analysis. The architecture integrates multiple trainable denoising modules within an end-to-end pipeline, reducing the need for manual intervention and enabling the automatic extraction of robust neural biomarkers. This design improves classification performance while enhancing interpretability and adaptability across subjects.
Domain adaptation techniques improve the effectiveness of spatiotemporal neural networks in EEG-based depression and epilepsy detection. For EEG-based depression detection, prior research struggles with (1) the reliance on extensive manual feature engineering to handle noise, (2) inadequate modeling of the spatial and temporal dynamics of brain activity, and (3) difficulty in adapting models to unseen patients. To address these challenges, we propose LAK-DSGCN (Lightweight Adjusted Kalman-aided Dual-Stream Graph Convolutional Networks), a novel spatiotemporal framework that (1) decomposes EEG signals into separate spatial and temporal components, (2) processes them using a gated TCN for temporal feature extraction and a GCN for spatial representation, and (3) fuses the learned representations using a lightweight Adjusted Kalman filter. Additionally, we incorporate a normalization term designed for the Kalman filter to enhance the model's generalizability across different patients.
For EEG-based epilepsy detection, existing approaches face three main limitations: (1) a reliance on high-quality, fixed-format EEG signals that do not account for real-world inconsistencies; (2) the inability to effectively handle irregular sampling rates, missing data, and noisy signals; and (3) a lack of robust feature-learning techniques to extract stable neural representations across patients. To address these issues, we introduce CPEDNet (Cross-Patient Epilepsy Diagnosis Network), which (1) employs a latent Neural Ordinary Differential Equation (NODE) module to enhance EEG signals by mitigating irregular sampling and missing data, (2) transforms EEG signals into brain network flow representations, capturing spatial-temporal dynamics, and (3) integrates a score-based self-supervised learning strategy to improve feature stability and cross-patient generalization.
Causal discovery techniques are applied to the task of drilling fluid loss prediction, which presents several unique challenges: (1) Data scarcity -- Due to the high cost of drilling operations, available datasets are often limited in size, increasing the risk of overfitting in causal models. (2) Complex causal structure -- Identifying robust causal relationships among covariates is difficult, yet essential for enabling generalization to unseen counterfactual scenarios based on causal reasoning. (3) Covariate influence -- It is nontrivial to ensure that different types of covariates influence the predicted fluid loss distribution in a causally consistent manner. To address these challenges, a causal discovery plug-in module is proposed for integration with Time Series Foundation Models (TSFMs). Specifically, the design provides three major contributions. (1) Frozen TSFM backbone -- The pretrained TSFM's parameters are frozen during fine-tuning to preserve general spatiotemporal representations and mitigate overfitting on small drilling datasets. (2) Causal rule integration -- Causal discovery techniques are used to identify and incorporate structured relationships between specific covariate subsets and the target variable, guiding prediction under counterfactual conditions. (3) Contrastive pretraining -- The plug-in module is pretrained using contrastive learning to ensure it learns discriminative latent representations conditioned on varying covariate configurations.
In summary, this dissertation advances the field of spatiotemporal data mining by addressing three core challenges—anomaly detection, domain adaptation, and causal discovery—that frequently hinder the robustness and generalization of existing models in real-world, cross-disciplinary scenarios.
We evaluate our proposed models on multiple real-world datasets, demonstrating significant improvements over existing state-of-the-art approaches in all settings. Together, these contributions push beyond the boundaries of conventional forecasting and classification tasks, demonstrating how task-specific adaptations and causal reasoning can greatly expand the applicability of spatiotemporal models in challenging, real-world environments.