University of Saskatchewan Research Archive
Not a member yet
14369 research outputs found
Sort by
Caring about the Extremes : Combining resources to calculate flood risk in the Bow River Basin
No full textCanada First Research Excellence FundNon-Peer ReviewedScientists, water managers, and engineers from all levels of government work together to predict flooding that threatens a major Canadian city
Radiation Phobia in Korea Provoked by the Fukushima Accident
No full textDr. Keon Wook Kang, a professor from the Department of Nuclear Medicine at Seoul National University Hospital, presented a paper titled "Radiation Phobia in Korea Provoked by the Fukushima Accident." He emphasized that the Fukushima nuclear accident has provoked fear of low-dose radiation, even among medical professionals. Some radiologists, who conduct ultrasonography, refused to examine patients who received PET/CT exams on the same day. Despite the estimated dose being well below 1 mSv per year, they declined to treat the patient, citing ALARA (As Low As Reasonably Achievable). This is more of an emotional reaction than a response based on scientific reasoning. Although people often believe that medical doctors are experts in radiation safety, this is not necessarily the case. As such, education on radiation safety should be reinforced in the medical school curriculum. Dr. Kang also addressed the fear surrounding seafood consumption in Korea, triggered by the plan to release treated water from Fukushima. Even though the estimated radiation dose from such seafood is negligible, public trust is undermined, in part, by experts who exaggerate the risk from such trivial doses. Governments, authorities, and radiation safety experts should consistently communicate with the public to debunk the myths and phobias that have been deeply entrenched by decades of media gaslighting
Applying Transformer-Based Deep Learning Model for Predicting Multimorbidity in Older Adults
No full textDisease predictive modelling supports decision-making for policymakers and healthcare providers, and aids in managing health conditions for individuals. Recent advances in deep learning have exhibited outstanding accuracy in diagnosis prediction. Bidirectional Encoder Representations from Transformers (BERT) has been applied to clinical research in disease prediction tasks due to its ability to comprehend entire diagnosis histories as sequences. However, there have been few studies conducted on the use of BERT models utilizing structured medical datasets for disease prediction. In particular, the increasing prevalence of multimorbidity is becoming a burden for the geriatric population, leading to adverse outcomes such as mortality, disability, and frequent healthcare utilization. Limited research exists that constructs Transformer models to predict multimorbidity, or that combines covariates in a multi-modal approach. This thesis aims to address these research gaps by constructing a Transformer pre-training model and fine-tuning model to predict multi-label outcomes of multimorbidity based on sequence-based diagnosis data and risk factors. Consequently, the pre-trained and fine-tuned models that can predict multimorbidity in the geriatric population were developed.
In this thesis, longitudinal data from the Korean Health Panel Survey (KHPS) was used for analysis. The data, collected by the Korea Institute for Health and Social Affairs and the National Health Insurance Service from 2008 to 2018, comprises up to 11 interviews. Older adults aged 60 or above (N=7,667) were selected for predicting multimorbidity, which was defined as having two or more chronic conditions. Age-related 60 chronic conditions were classified, and the twelve most prevalent chronic diseases were selected for prediction: hypertension, musculoskeletal and joint diseases, inflammatory arthropathies, diseases of the esophagus/stomach/duodenum, dorsopathies, diabetes, ear/nose/throat diseases, dyslipidemia, osteoporosis, eye diseases, peripheral neuropathy, and colitis/related diseases. Explanatory variables-such as sex, age, Body Mass Index (BMI), life insurance, and income quintile-were selected based on their significance in the multivariable Generalized Estimating Equations (GEE) model. Pre-training was conducted by building a Masked Language Model (MLM) to capture disease representations. The Medical Information Mart for Intensive Care (MIMIC-III) dataset was employed as an external validation dataset to test the pre-trained model. Metrics such as precision, recall, F1-score, average precision, and the Area Under the Receiver Operating Characteristic curve (AUROC) were used to present experimental results. The fine-tuned model and the scratch-trained model were compared using both the KHPS and MIMIC-III datasets. In the fine-tuning stage, multi-label classification was conducted to predict multiple chronic conditions 1 month and 1 year after a certain follow-up point in older adults. The performance of the fine-tuned model was compared to machine learning models such as binary relevance, classifier chain, and label powerset. Lastly, our final model was compared to existing Transformer multi-label classification models, BEHRT and Med-BERT.
The use of a pre-trained model improved macro precision by 7.9% and macro recall by 1.3% compared to the scratch-trained model when the validation dataset, MIMIC-III, was utilized. Incorporating variables combination exhibited no significant differences. Macro precision values for Exp1 (sex, age, BMI, and income), Exp2 (sex, age, and BMI), and Exp3 (sex and age) were 0.431, 0.428, 0.436, and 0.431, respectively, indicating minimal differences. The binary relevance achieved approximately 9.3% higher macro precision but exhibited a 24.8% lower macro recall compared to our model. When compared to the BEHRT model, our model had 12.6% lower macro precision and 15.1% higher macro recall.
These findings indicate that a pre-training model has the potential to improve a model's performance when applied to datasets with similar characteristics. Overall, our model exhibited a higher recall than the multi-label machine learning methods and the deep Transformer models, BEHRT and Med-BERT. Furthermore, this thesis provided a model capable of predicting multimorbidity 1 month and 1 year after a certain follow-up point. With robust validation and improved precision, recall, and F1-score, our model can assist both individuals and medical experts in managing health conditions and medical costs in the future
A New Family of High-Current Cyclotrons
No full textWe have developed a new family of compact cyclotrons capable of 10 mA of protons at energies up to 60 MeV. Designed for the IsoDAR neutrino experiment [1], the x10 current increase could find important applications in isotope production, for products with long half-lives (68Ge) or low cross sections (232Th(p,X)225Ac). A recent development has been use of the breakup of 40-MeV deuteron ions forming an intense fast-neutron field suitable for 226Ra(n,2n)225Ra --> 225Ac, a highly-efficient channel for 225Ac production. Achieving and Using High Currents Beam current limits on today’s isotope cyclotrons arise from extraction foil lifetime and central region erosion from inefficient injection. Our cyclotrons accelerate H2+ ions, reducing the effect of space charge, and bunch 90% of the beam into the RF acceptance window with an RFQ. A beam-dynamics effect discovered at PSI called “vortex motion” translates space charge forces in the cyclotron magnetic field into lateral motion that stabilizes the individual RF beam packets into compact bunches. Halo is generated but is collimated in the first turns. Stable packets are formed in about the first few MeV (~5 turns), allowing for adequate turn separation to enable clean septum extraction of the H2+ ions without use of a stripper foil. Using these principles we can build cyclotrons for energies ranging from a few MeV up to 60 MeV. Turn separation above 60 MeV is lower, but use of structure resonances may enable extending the energy to higher values. Q/A of H2+ is 0.5, so cyclotrons are suitable for D+, He+, C6 ions at the same energy/nucleon. At present, the H2+ ion source and RFQ are ready, the 1.5 MeV Demonstrator is being built, the IsoDAR experiment will be deployed in 5 years. x10 current means x10 beam power. Developing high-power targets will enable the most efficient utilization of the available beam. In the interim, splitting the beam onto as many as 10 target stations is feasible, by employing either RF kicking of bunches into separate beam lines, and/or by insertion of stripper foils into the edges of the extracted H2+ beam to peel off adjustable amounts of protons using a dipole right after the stripper. Conclusions Our designs represent a paradigm shift in cyclotron performance, shattering the beam-current barrier, and opening the door for novel applications. In addition to isotope production, these machines are compact high-flux MeV-range neutron generators; planned uses extend from neutrino production to cost-effective IFMIF-style fusion reactor materials testing platforms. References 1 J.Alonso et al (2022) Neutrino Physics Opportunities with the IsoDAR Source at Yemilab. Phys. Rev. D, 105, 075150. https://doi.org/10.1103/PhysRevD.105.05200
The Impact of Garlic-infused Supplements on Beef Cattle Health and Performance
No full textA 2-year study was conducted to investigate the effect of free-choice garlic-infused supplements on the performance and health of feedlot beef steers. Collection of performance data included diet and mineral intake, fly abundance, average daily gain (ADG), liver abscess scores and carcass quality. Animal health was assessed through blood components, fecal parasite load and short-chain fatty acids (SCFA) profiles. A total of 208 yearling steers (mean BW±SD; 511 ± 42 kg) were randomly assigned to 1 of 4 treatments, either non-garlic mineral supplement (MS) only, MS + 0.3% garlic oil-based premix (0.3GO), MS + 2.5% garlic powder (2.5GP) or MS+ 5% garlic powder (5GP). Animals were fed a high-grain diet and mineral supplement ad libitum in separate GrowSafe™ bunks to measure individual intake quantity and feeding behavior. Feeding trials were 86 d and 108 d in Yr 1 and 2, respectively.
Data from GrowSafe™ bunks showed treatments did not cause difference in feed intake but there was an increase in supplement intake in all three treatment groups (P 0.05) were observed for fly abundance between treatment groups, possibly due to recent insecticide application and low fly hatch rate in frequently disturbed grain-based manure. Likewise, no significant differences (P > 0.05) were observed for ADG, complete blood count, parasite load, liver abscess scores or carcass quality parameters among treatment groups.
Study results suggest garlic treatment showed no performance benefit or adverse health impact in growing steers. This study revealed the addition of garlic products can potentially increase mineral supplement consumption which may be used as a feeding strategy to stimulate intake. Future research is required to further clarify whether variability in allicin content is responsible for the inconsistency seen between the two years. We recommend that future studies extend the duration of feeding trials for more comprehensive assessment of the long-term effects of garlic. Additionally, housing animals in pasture settings would enhance fly abundance, providing a more accurate evaluation of fly control measures
Branding Athletics in Canadian Higher Education
No full textHigher education institutions require competitive branding strategies to differentiate product
offerings in the higher education market. Higher education institutions incorporate multiple
brand identities into branding strategies that offer unique offerings to various audiences.
Extending the brand identity of an institution is a common branding strategy, allowing a deeper,
more meaningful connection with specific audiences internally and externally. A successful
extension of higher education institutions is the introduction of an athletic program for collegiate
sport competition. The athletic program often develops an alternative brand identity differing
from the institution identity. The scope of this thesis considers how athletic programs are
branded and named at Canadian higher education institutions and what impact this has on brand
alignment with the overall institution brand identity
Towards reducing the high cost of parameter sensitivity analysis in hydrologic modelling: a regional parameter sensitivity analysis approach
No full textAccepted manuscript submitted to HESS.Canada First Research Excellence FundPeer ReviewedLand surface models have many parameters that have a spatially variable impact on model outputs. In applying these models, sensitivity analysis (SA) is sometimes performed as an initial step to select calibration parameters. As these models are applied on large domains, performing sensitivity analysis across the domain is computationally prohibitive. Here, using a VIC deployment to a large domain as an example, we show that watershed classification based on climatic attributes and vegetation land cover helps to identify the spatial pattern of parameter sensitivity within the domain at a reduced cost. We evaluate the sensitivity of 44 VIC model parameters with regard to streamflow, evapotranspiration and snow water equivalent over 25 basins with a median size of 5078 km2 15 . Basins are clustered based on their climatic and land cover attributes. Performance of transferring parameter sensitivity between basins of the same cluster is evaluated by the F1 score. Results show that two donor basins per cluster are sufficient to correctly identify sensitive parameters in a target basin, with F1 scores ranging between 0.66 (evapotranspiration) to 1 (snow water equivalent). While climatic attributes are sufficient to identify sensitive parameters for streamflow and evapotranspiration, including vegetation class significantly improves skill in identifying sensitive parameters for snow water equivalent. This work reveals that there is an opportunity to leverage climate and land cover attributes to greatly increase the efficiency of parameter sensitivity analysis and facilitate more rapid deployment of land surface models over large spatial domains
Complex Interplay of Mercury and Arsenic with Sulfur and Selenium in Biological Systems
No full textThe abstract of this item is unavailable due to an embargo
DOMESTIC PRODUCTION OF Mo-99 AND Ac-225 USING COMMERCIAL PWR AND FAST EXPERIMENTAL REACTOR JOYO IN JAPAN
No full textThe production technology of medical radioisotopes (RI) using existing nuclear fission reactors has been studied to improve/achieve their domestic preparedness in Japan. The target nuclides currently considered in our project are Mo/Tc which is the most commonly used ones in medical diagnosis and Ac-225 which is recently known as effective alpha emitting nuclide for targeted alpha-particle therapy. Existing fission reactors, PWRs and Joyo in Japan, have potentials to work as excellent facilities for medical isotope production, as by-products of heat/electricity generation without consuming electricity and needs for new plant construction