Scholarship @ UWindsor

Scholarship @ UWindsor is the institutional repository of the University of Windsor (UWindsor), showcasing and preserving the UWindsor community’s scholarly outputs, as well as items from the Leddy Library’s Archives & Special Collections. Its mission is to disseminate and preserve knowledge created or housed at the University of Windsor.

Contact scholarship@uwindsor.ca for more information.

Communities in Scholarship @ UWindsor

Select a community to browse its collections.

Now showing 1 - 4 of 4

Conferences and Conference Proceedings 3167
Papers, presentations and abstracts of conferences held at the University of Windsor, in person and virtually.
Digitized Local Collections 1552
Digitized local items from the collections of the Leddy Library, University of Windsor, and community partners.
Faculties, Departments and Research Units 3171
Open Access Faculty publications, reports and working papers from academic departments at the University of Windsor.
Theses, Dissertations, and Major Papers 9577
Formal graduate original research from the University of Windsor's Masters and Doctoral programs.

Recent Submissions

Access status: Open Access ,
Epitope mapping of a neutralizing antibody against rabbit hemorrhagic disease virus GI.2
Podadera, Ana; Leuthold, Mila; Martín-Alonso, José Manuel; Casais, Rosa; Álvarez, Angel Luis; Lobo-Castañón M.J.; Parra, Francisco; Dalton, Kevin Paul
In 2010, rabbit hemorrhagic disease virus (RHDV) GI.2 emerged, and unlike RHDV GI.1, it caused mortality in young rabbits, while existing vaccines were not fully protective. The GI.2-specific monoclonal antibody (mAb) 2D9 has been used as a tool to discriminate between these viruses in diagnostic tests. In this study, we mapped the binding epitope for 2D9 on the GI.2 The VP60 capsid protein demonstrated the neutralizing capacity of this mAb, which was able to prevent GI.2 infections in an experimental challenge. Our results suggest that external loops (1, 4 and 5) in the P2 subdomain of VP60 contribute to the discontinuous neutralizing epitope recognized by mAb 2D9. Moreover, analysis of naturally occurring RHDV GI.2 isolates revealed key residues involved in mAb 2D9 binding that are under selective pressure. The findings described in this work provide valuable information regarding our understanding of virus neutralization and immune escape, which may help in the development of novel antiviral compounds.
Access status: Open Access ,
Measuring the fitted filtration efficiency of cloth masks, medical masks and respirators
(PLoS, 2025-04-21) Tomkins, Amanda A. a ; a ; b ; a ;; Dulai, Gurleen; Dulai, Ranmeet; Rassenberg, Sarah; Lawless, Darren; Laengert, Scott; Hasan, Shiblul; de Lannoy, Charles-Francois; Drouillard, Ken G.; Clase, Catherine M.
Masks reduce transmission of SARS-CoV2 and other respiratory pathogens. Comparative studies of the fitted filtration efficiency of different types of masks are scarce. Objective To describe the fitted filtration efficiency against small aerosols (0.02–1 µm) of medical and non-medical masks and respirators when worn, and how this is affected by user modifications (hacks) and by overmasking with a cloth mask. Design We tested a 2-layer woven-cotton cloth mask of a consensus design, ASTM-certified level 1 and level 3 masks, a non-certified mask, KF94s, KN95s, an N95 and a CaN99. Setting Closed rooms with ambient particles supplemented by salt particles. Participants 12 total participants; 21–55 years, 68% female, 77% white, NIOSH 1–10. Main Outcome and Measure Using standard methods and a PortaCount 8038, we counted 0.02–1 µm particles inside and outside masks and respirators, expressing results as the percentage filtered by each mask. We also studied level 1 and level 3 masks with earguards, scrub caps, the knot- and-tuck method, and the effects of braces or overmasking with a cloth mask. Results Filtration efficiency for the cloth mask was 47–55%, for level 1 masks 52–60%, for level 3 masks 60–77%. A non-certified KN95 look-alike, two KF94s, and three KN95s filtered 57–77%, and the N95 and CaN99 97–98% without fit testing. External braces and overmasking with a well-fitting cloth mask increased filtration, but earguards, scrub caps, and the knot-and-tuck method did not. Limitations Limited number of masks of each type sampled; no adjustment for multiple comparisons. Conclusions and Relevance Well-fitting 2-layer cotton masks filter in the same range as level 1 masks when worn: around 50%. Level 3 masks and KN95s/KF94s filter around 70%. Over a level 1 mask, external braces or overmasking with a cloth-mask-on-ties produced filtration around 90%. Only N95s and CaN99s, both of which have overhead elastic, performed close to the occupational health and safety standards for fit tested PPE (>99%), filtering at 97–99% when worn, without formal fit testing. These findings inform public health messaging about relative protection from aerosols afforded by different mask types and explain the effectiveness of cloth masks observed in numerous epidemiologic studies conducted in the first year of the pandemic. A plain language summary of these findings is available at https:// maskevidence.org/masks-compared.
Access status: Open Access ,
A Coupled Multiphysics Framework for Advanced Characterization of PWM-Driven Induction Motors
(IEEE) Taqavi, Omolbanin; Song, Pengzhao; Bourgault, Alexandre J.; Li, Ze; Byczynski, Glenn; Kar, Naryan C.
The design of high-performance electrical machines necessitates the intricate integration of multiple physical domains, including electromagnetic, mechanical, and thermal aspects. To meet the ever-evolving demands of the field, there is a pressing need for a platform capable of simultaneously analyzing these interconnected phenomena with both precision and efficiency, all within a practical premanufacturing timeframe. This paper introduces a semi-analytical multiphysics assessment framework tailored for inverter-fed traction induction machines (IMs). By integrating three core physical domains, the framework leverages an electromagnetic model to evaluate rotor and stator currents, traction characteristics, and radial air-gap flux density. Vibroacoustic models are employed to predict noise and vibration induced by electromagnetic forces, while a three-dimensional (3D) nodal network thermal model captures transient temperature distributions across motor components. These highly efficient models are seamlessly integrated into a unified framework, allowing for a thorough and precise analysis of any IM in an efficient and timely manner. The framework is also designed for easy integration with optimization tools, enhancing its applicability for performance and design optimization. The developed scheme is examined on an enclosed IM prototype and is verified by comprehensive finite element analyses and experimental testing. The findings introduce a novel approach that integrates advanced computational tools with traditional design methods to assess the multiphysics performance of IMs across their entire performance spectrum. The developed method holds applicability across various applications with implications for other machine types.
Access status: Open Access ,
A Multistage Detection Framework Based on TFA and Multiframe Correlation for HFSWR
(IEEE, 2025-01-09) Li, Zongtai; Li, Gangsheng; Zhang, Ling; Liu, Lanjun; Wu, Q. M. Jonathan
Maritime surveillance heavily relies on high-frequency surface wave radar (HFSWR) systems. However, clutter and interference make it difficult to accurately detect vessel targets using a single-frame detection method. This study introduces an improved time-frequency analysis (TFA) algorithm to enhance the features in single-frame detection. In this article, TFA, multiframe correlation, and deep neural networks are integrated to develop a three-stage detection framework. First, faster R-CNN is customized for the preprocessing stage to identify sea clutter regions. Then, based on the range-Doppler (RD) spectrum, suspicious targets are swiftly identified amidst clutter in the initial stage. Subsequently, the improved TFA algorithm is applied to adjacent range cells of suspicious targets to generate multiframe TF images, forming a three-dimensional data block structured as time-RD frequency. To reduce computational complexity, a TFA method using multisynchrosqueezing transform is employed, enhancing detection accuracy for targets within cluttered regions. In the final stage, a 3DResnet model is utilized to leverage the differences in features between clutter and targets across three dimensions. This allows for distinguishing genuine targets from false targets using time series information from multiple frames. Comparative analysis against classical target detection algorithms demonstrates the superior detection performance of the proposed framework within clutter regions. This showcases its potential for enhancing the maritime surveillance capabilities of HFSWR.
Access status: Open Access ,
PolyLLM: polypharmacy side effect prediction via LLM-based SMILES encodings
(Frontiers Media SA, 2025-07-31) Hakim, Sadra; Ngom, Alioune
Polypharmacy, the concurrent use of multiple drugs, is a common approach to treating patients with complex diseases or multiple conditions. Although consuming a combination of drugs can be beneficial in some cases, it can lead to unintended drug-drug interactions (DDI) and increase the risk of adverse side effects. Predicting these adverse side effects using state-of-the-art models like Large Language Models (LLMs) can greatly assist clinicians. In this study, we assess the impact of using different LLMs to predict polypharmacy. First, the chemical structure of drugs is vectorized using several LLMs such as ChemBERTa, GPT, etc., and are then combined to obtain a single representation for each drug pair. The drug pair representation is then fed into two separate models including a Multilayer Perceptron (MLP) and a Graph Neural Network (GNN) to predict the side effects. Our experimental evaluations show that integrating the embeddings of Deepchem ChemBERTa with the GNN architecture yields more effective results than other methods. Additionally, we demonstrated that utilizing complex models like LLMs to predict polypharmacy side effects using only chemical structures of drugs can be highly effective, even without incorporating other entities such as proteins or cell lines, which is particularly advantageous in scenarios where these entities are not available.