HARVEST
University of Saskatchewan's Repository for Research, Scholarship, and Artistic Work
Welcome to HARVEST, the repository for research, scholarship, and artistic work created by the University of Saskatchewan community. Browse our collections below or find out more and submit your work.
Communities in HARVEST
Select a community to browse its collections.
Recent Submissions
VernierVision: Automatic Reading of Vernier Scales for Historic Observatory Telescopes
(2025-03-18) Wheler, Timothy Jeffrey; Stavness, Ian; Stavness, Ian; Gutwin, Carl; Eramian, Mark
Advancements in deep space optics have revolutionized astronomical research, rendering older observa-
tory technologies obsolete. However, these historical resources still hold significant potential to inspire future
stargazers. Historic observatories, although outdated for contemporary research, are invaluable for students
and amateur astronomers constrained by the cost of modern equipment. This thesis aims to integrate elec-
tronic measurement systems into historical telescopes to enable remote and automated use while preserving
their original condition.
Specifically, the thesis focuses on the Observatory at the University of Saskatchewan, a landmark facility
that remains a landmark of substantial significance. Modernizing these telescopes without damaging their
irreplaceable components is crucial. Many historical telescopes use manual rotary Vernier scales to measure
direction, unlike modern telescopes with rotary encoders. The telescope in this study uses Vernier scales
for right ascension and declination measurements. Instead of replacing these scales, this thesis devises and
evaluates computer vision approaches to automatically read the original Vernier scales.
A computer vision system comprising cameras, object/line detection, and post-processing pipelines was
developed to automate scale measurements. The system unifies numeral localization and classification using
an object detection network and creates a robust post-processing pipeline for region-of-interest (ROI) and
feature extraction. The performance of the final system was sufficient for one of the two scales, however due
to additional sources of error, the performance on the second scale was lacking for high precision operation
High-Sensitivity and Flexible Motion Sensing Enabled byRobust, Self-Healing Wood-Based Anisotropic Hydrogel Composites
(Small, 2025-02) Teng, Youchao; Zhang, Zhilei; Cui, Yunqi; Su, Zhe; Godwin, Matthew; Chung, TzuChun; Zhou, Yongzan; Leontowich, Adam F. G.; Islam, Muhammad Shahidul; Tam, Kam C.; Wu, Yimin A.
By integrating polyvinyl alcohol (PVA)-borate-tannic acid (TA)-sodium sulfate into cellulosic wood matrices, a novel wood-basedPVA-borate-TA-sodium sulfate (WPBTS) hydrogel is successfully synthesized. Through a multicomponent synergistic design combining natural lignocellulose, PVA, borax, TA, and sodium sulfate, multiple dynamic cross-linking mechanisms—dynamic borate bonding, hydrogen bonding, and metal-ligand interactions—are established, resulting in WPBTS hydrogels with exceptional mechanical properties and self-healing capabilities. The mechanical strength of the WPBTS hydrogel reached an impressive 19.8 MPa, a 45-fold increase compared to PVA-borax-tannic acid (PBTS) hydrogels. Furthermore, the assembled WPBTS hydrogel-based flexible sensor demonstrates a remarkably fast response time of just 20 ms and maintains excellent performance in challenging simulated saline environments. This innovation represents a significant advancement in sensor technology and highlights the potential for transformative applications in complex and demanding scenarios.
Machine learning models for predicting interaction affinity energy between human serum proteins and hemodialysis membrane materials
(Scientific Reports, 2025-01) Nazari, Simin; Abdelrasoul, Amira
Membrane incompatibility poses significant health risks, including severe complications and potential fatality. Surface modification of membranes has emerged as a pivotal technology in the membrane industry, aiming to improve the hemocompatibility and performance of dialysis membranes by mitigating undesired membrane-protein interactions, which can lead to fouling and subsequent protein adsorption. Affinity energy, defined as the strength of interaction between membranes and human serum proteins, plays a crucial role in assessing membrane-protein interactions. These interactions may trigger adverse reactions, potentially harmful to patients. Researchers often rely on trial-and-error approaches to enhance membrane hemocompatibility by reducing these interactions. This study focuses on developing machine learning algorithms that accurately and rapidly predict affinity energy between novel chemical structures of membrane materials and human serum proteins, based on a molecular docking dataset. Various membrane materials with distinct characteristics, chemistry, and orientation are considered in conjunction with different proteins. A comparative analysis of linear regression, K-nearest neighbors regression, decision tree regression, random forest regression, XGBoost regression, lasso regression, and support vector regression is conducted to predict affinity energy. The dataset, comprising 916 records for both training and test segments, incorporates 12 parameters extracted from data points and involves six different proteins. Results indicate that random forest (R² = 0.8987, MSE = 0.36, MAE = 0.45) and XGBoost (R² = 0.83, MSE = 0.49, MAE = 0.49) exhibit comparable predictive performance on the training dataset. However, random forest outperforms XGBoost on the testing dataset. Seven machine learning algorithms for predicting affinity energy are analyzed and compared, with random forest demonstrating superior predictive accuracy. The application of machine learning in predicting affinity energy holds significant promise for researchers and professionals in hemodialysis. These models, by enabling early interventions in hemodialysis membranes, could enhance patient safety and optimize the care of hemodialysis patients.
The Instrumentalization of Drugs in the Hannibal Lecter Tetralogy
(2025-03-17) Yang, Haonan; Banco, Lindsey; Martin, Ann; Liu, Yin
Thomas Harris’s Hannibal Lecter novels create “Hannibal the Cannibal”—an icon in modern pop culture and a controversial character who “transgresses all categories—culture/ nature, crime/ order, human/ animal, rationality/ madness” (Mizejewski 175). In this paper, I will adopt comparative and historicist approaches to explore the representation of drugs in the Hannibal Lecter tetralogy to discern how these texts understand the relationship between good and evil. In the tetralogy, Dr. Lecter instrumentalizes drugs for three purposes—vengeance, playing God, and healing traumas. Lecter’s instrumentalization of drugs not only transforms him into a complicated cannibalistic serial murderer who kills for justice but also transforms Starling into his romantic partner. Drawing on Jacques Derrida’s concept of pharmakon, I will argue that Lecter’s utilization of drugs in his life journey demonstrates the dual nature of his humanity. His instrumentalization of drugs as pharmaka complicates the conventional understanding of drugs in crime fiction primarily as a means of pursuing recreational intoxication and a tool for criminals to achieve evil purposes. Harris’s portrayal of instrumentalizing drugs suggests that psychoactive drugs help to represent the duality of humanity, and, for fictional serial killers, drugs are not a means of self-intoxication but have the multifarious functions of assault, self-protection, and dealing with traumas. By associating cannibalistic serial killing with drug use, Harris critiques the problematic medical system in America, the corruption of the American judicial system, and the theme of excess in American consumption culture.
The hysteretic and gatekeeping depressions model − A new model for variable connected fractions of prairie basins
(Elsevier, 2025-06) Shook, Kevin; Pomeroy, John W.
The Prairie Pothole Region of western North America has unusual hydrology and hydrography. Its level, post-glacial topography means that many drainage basins are dominated by internally drained depressions, rather than having conventional dendritic drainage networks of stream channels. Modelling the hydrology of these regions is difficult because the relationship between depressional storage and the connected fraction of a basin is hysteretic. Existing models are either computationally intensive and require high-resolution Digital Elevation Model (DEM) data which may not exist or require calibration and cannot reproduce the hysteresis between the basin connected fraction and depressional storage. The Hysteretic and Gatekeeping Depressions Model (HGDM) has been developed to simplify modelling of prairie basins with variable connected/contributing fractions. The model uses “meta” depressions to model the hysteretic responses of small depressions and a discrete model of large depressions, which cause “gatekeeping”, meaning that they prevent upstream flows from reaching the outlet until the depressions are filled. The HGDM was added to the Cold Regions Hydrological Modelling (CRHM) platform which is one of the few models that has successfully simulated land surface hydrology in the Canadian Prairies. CRHM + HGDM is tested by modelling streamflows at Smith Creek, a basin in southeastern Saskatchewan, Canada. It is demonstrated that CRHM + HGDM can reproduce the relationship between the connected/contributing fractions of sub-basins and their depressional storage at least as well as existing models. Importantly, it appears that HGDM can be used with coarse-resolution DEMs, which may permit its use in the many locations where higher-resolution data is unavailable. The simplicity and limited parameterization needs of HGDM may allow for broader representation of depressions and variable contributing area in prairie hydrology.
Deep Learning Methods for Cervical Cell Classification with Cytology Image Data
(2025-03-14) Fang, Ming; Bui, Francis francis.bui@usask.ca; Ko, Seok-Bum; Vizeacoumar, Franco; Wahid, Khan; Maleki, Farhad
Advances in biomedical imaging modalities can provide rich information about the properties of normal and cancerous tissues, assisting medical experts in examining and interpreting the current status of cancers. Biomedical image classification plays an integral role in the diagnosis and treatment of various cancers. Traditional methods have some disadvantages, such as time-consuming and labor-intensive, which significantly affect the effective detection of cancer. In response to these issues, it is necessary to apply computer-aided techniques to automatically analyze medical images, so that patients can be diagnosed and treated timely. With the success of deep learning techniques in the field of computer vision, more and more researchers choose machine and deep learning models as major tools for analyzing biomedical images. Leveraging some machine learning-based algorithms, e.g., classification algorithms, is helpful for medical experts to diagnose disease status since computers may discover the patterns that are difficult for humans to see.
Cervical cancer is seriously threatening the lives of women around the world. Fortunately, cervical cancer can be detected and treated as long as it is diagnosed at the early stage. Although traditional methods such as Pap smear test can reduce the death rate of cervical cancer, they are time-consuming and prone to errors. With the emergence of machine and deep learning algorithms, cervical cancer has been diagnosed and treated effectively. However, there are still some challenges in the current detection methods. First, the importance of features of various levels is not fully taken into account in cytology image analysis. Second, the majority of relevant studies only pay attention to the local detailed information while ignoring the global feature information of cells. Moreover, most researchers only focus on the features of images themselves, rather than involving the properties of cervical cells.
To improve these issues, I have developed novel and effective computational methods in Chapters 3-5 of this thesis to automatically analyze cervical cell images, further assisting in the diagnosis and treatment of early cervical cancer.
Chapter 3 develops a new deep convolutional neural network (CNN) that learns feature patterns by applying multiple kernels with different sizes to automatically classify cervical cells from cytology images. Three different basic modules are designed to capture feature information through multiple kernels with different sizes. Afterwards, several such basic modules are stacked to form the cervical cell classification model. Chapter 4 considers the importance of global features. A deep integrated feature fusion module is proposed to synergize local and global cell features more effectively. Based on a parallel architecture with a CNN branch and a transformer branch, the local and global features are extracted simultaneously, enhancing the discriminative ability of cell features. Chapter 5 incorporates supervised contrastive learning with CNN to construct a novel cervical cell classification framework, to guide the deep model to acquire better cell representations. By generating multiple views of cell images with data augmentation to simulate the pattern variations of cervical cells and jointly training the model with our designed hybrid loss, the proposed framework improves the robustness in cervical cell representations and addresses challenges, including the pattern variations in cervical cells caused by the data acquisition process and the misclassification of cells with similar pathological properties.
In summary, this thesis begins with relevant background knowledge that describes the proposed research objectives and pipelines. Subsequently, the thesis presents a comprehensive review of the existing literature on deep learning-based cervical cell image analysis, identifying key findings and challenges in the current research, as well as providing a foundation for the subsequent analysis. Next, this thesis develops several new methodologies for automatic detection of cervical cell abnormalities. A substantial number of comparative experiments demonstrate the effectiveness of the developed methods, which could be viewed as an auxiliary tool to support clinical use for cervical cancer.
Geochemical Evolution of Froth Treatment Tailings at and Oil Sands Mine in Northern Alberta, Canada
(2025-03-14) Paulsen, Joshua J; Lindsay, Matthew B.J.; Ferguson, Grant; Smith, Laura A; Sproat, Colin D
Froth Treatment Tailings (FTT) are generated during water-based bitumen extraction in the oil sands mining operations in Northern Alberta, Canada. Acid generation and metal(loid) release during sulfide-mineral oxidation is an emerging consideration for management and reclamation of FTT deposits. This pyrite-bearing tailings stream is deposited sub-aerially where post-depositional hydraulic segregation of FTT slurry generates beach deposits susceptible to oxidative weathering. Recent studies have demonstrated the potential for acid generation and metal(loid) release during oxidative weathering of FTT however, the geochemical evolution of commercial scale FTT beach deposits is not fully understood. Refining the geochemical conceptual model of FTT beach deposits will provide insight into tailings management strategies and landform reclamation planning. This research examines the geochemical evolution of a commercial-scale sub-aerial FTT beach deposit.
A sampling campaign was completed on the Plant 6 FTT beach deposit at the Mildred Lake mine operated by Syncrude Canada Ltd (Syncrude). Continuous core sampling was conducted on five boreholes collected along a transect of the P6 FTT Beach deposit which approximately followed historic FTT deposition points. Boreholes were drilled to a depth ranging from approximately 30 to 46 m and tailings solid samples were collected along the borehole depth profiles at each location. Tailings pore-water was extracted by mechanical squeezing from solid samples and tailings solids were sub-sampled for subsequent analysis.
Mineralogical analysis revealed that FTT beach solids are generally dominated by quartz plus various phyllosilicates, sulfides, carbonates, oxides, phosphates, and additional silicates occurring at minor to trace amounts. Sulfide and carbonate mineral contents each range from approximately 5 to 10 wt. % but their abundance can exceed 20 wt. % at discrete locations. Pyrite and minor marcasite are the only sulfides detected in FTT beach samples. Siderite is the dominant carbonate in both weathered and non-weathered FTT beach samples, which commonly contain lesser amounts of dolomite, ankerite, and calcite. Pore-waters in non-weathered regions of the FTT beach deposit are generally characterized by circum-neutral pH (7–8) and relatively low dissolved SO4, Fe, and metal(loid) concentrations.
Vegetable and fruit intake and diet quality changes in women who participate in prenatal programming involving a dietitian: A scoping review.
(2025-03-14) Boyer, Audrey; Lieffers, Jessica; Cammer, Allison; Jorgensen, Derek; Vatanparast, Hassan
The abstract of this item is unavailable due to an embargo.
Data-driven exploration of weak coordination microenvironment in solid-state electrolyte for safe and energy-dense batteries
(Springer Nature, 2025) Lao, Zhoujie; Tao, Kehao; Xiao, Xiao; Qu, Haotian; Wu, Xinru; Han, Zhiyuan; Gao, Runhua; Wang, Jian; Wu, Xian; Chen, An; Shi, Lei; Chang, Chengshuai; Song, Yanze; Wang, Xiangyu; Li, Jinjin; Zhu, Yanfei; Zhou, Guangmin
The unsatisfactory ionic conductivity of solid polymer electrolytes hinders their practical use as substitutes for liquid electrolytes to address safety concerns. Although various plasticizers have been introduced to improve lithium-ion conduction kinetics, the lack of microenvironment understanding impedes the rational design of high-performance polymer electrolytes. Here, we design a class of Hofmann complexes that offer continuous two-dimensional lithium-ion conduction channels with functional ligands, creating highly conductive electrolytes. Assisting with unsupervised learning, we use Climbing Image-Nudged Elastic Band simulations to screen lithium-ion conductors and screen out five potential candidates that elucidate the impact of lithium coordination environment on conduction behavior. By adjusting the covalency competition between Metal−O and Li−O bonds within Hofmann complexes, we can manipulate weak coordination environment of lithium-ion for rapid conduction kinetics. Li | |sulfurized polyacrylonitrile (SPAN) cell using solid-state polymer electrolytes with predicted Co(dimethylformamide)2Ni(CN)4 delivers an initial discharge capacity of 1264 mAh g−1 with a capacity retention of 65% after 500 cycles at 0.2 C (335 mA g−1), at 30 °C ± 3 °C. The assembled 0.6 Ah Li | |SPAN pouch cell delivers an areal discharge capacity of 3.8 mAh cm−2 at the second cycle with a solid electrolyte areal mass loading of 18.6 mg cm−2 (mass-to-capacity ratio of 4.9).
Enhancing drying efficiency and terpene retention of cannabis using cold plasma pretreatment
(Elsevier, 2025-02-11) Das, Pabitra Chandra; Heydari, Mohamad Mehdi; Baik, Oon-Doo; Zhang, Lifeng; Tabil, Lope
Hang-drying of cannabis at room conditions is a slow process and leads to the risk of microbial growth. This method can sometimes prevent cannabis from reaching the equilibrium moisture content (EMC) below the safe storage threshold. On the other hand, high-temperature drying techniques are faster but negatively impact the secondary metabolites. Cold plasma (CP) is a novel technique explored in this study to treat cannabis at various operational conditions of plasma jet (power: 300, 350, and 400 W, time: 20, 30, and 40 s) prior to drying at environmental conditions of 25°C and 50 % RH. The findings revealed that untreated cannabis samples reached an equilibrium moisture content (EMC) of approximately 16 % in 1260 min. In contrast, CP-pretreated samples achieved lower EMCs of 10–14 % within 690–840 min. CP pretreatment also resulted in high moisture diffusivity, lower energy consumption, and higher energy efficiency. Increasing CP power and residence time accelerated the decarboxylation of cannabinoids, leading to the formation of more tetrahydrocannabinol (THC) and less tetrahydrocannabinolic acid (THCA), without significantly affecting the total THC (27.45 % untreated vs. 25.82 % - 28.36 % g/ g of dry matter in CP pretreated samples). Compared with untreated dried inflorescences, the 400 W and 30 s CP treated inflorescences resulted in the retention of 96 % of terpenes, whereas all 300 W CP treated samples retained > 90 % of terpenes. Overall, the study highlights that CP pretreatment is a promising technology for the cannabis industry in shortening the drying time and preserving the product quality, especially terpenes.