39339 research outputs found
Sort by
Evaluating the MedMira Multiplo® Complete Syphilis (TP/nTP) antibody test in a sexually transmitted infection clinic in Ottawa, Canada: increased rapid diagnosis and improved antibiotic stewardship
No full textAbstract Background Syphilis now affects every population and serology is the mainstay of diagnosis. The issue is that serology has a turnaround time of several days. One solution is point-of-care tests (POCTs), which can provide results in minutes. We consequently evaluated the MedMira Multiplo® Complete Syphilis Test in an STI clinic in Ottawa, Canada. Methods Anyone 16 + years old who consented and was undergoing syphilis testing at our clinic was eligible. Those who enrolled completed the POCT and saw a clinician to review their result. We calculated sensitivities and specificities for the POCT, compared to serology and diagnosis. Results From August 2024 to May 2025, we performed 622 syphilis POCTs on 600 participants. Compared to serology when chemiluminescent microparticle immunoassay (CMIA) and Treponema pallidum particle agglutination (TP.PA) tests were reactive, the POCT treponemal (TP) test had a sensitivity of 90.1% and specificity of 97.9%. Compared to any dilution of rapid plasma reagin (RPR), the POCT non-treponemal (nTP) test had a sensitivity of 82.5% and specificity of 99.1%. When we stratified POCT nTP results based on RPR titers, the POCT nTP had a sensitivity of 94.1% for RPR dilutions ≥ 1:8. Compared to serology, the POCT identified 91.4% of new syphilis infections and 97% of infectious syphilis. Conclusions POCTs informed clinical syphilis management. While most research has focused on how POCTs can facilitate treatment, in our study, there was a second major utility: to withhold antibiotics when recommended as empiric treatment but when the patient does not have active syphilis. Future research on syphilis POCTs should focus on their abilities to rule in and rule out infections. Trial registration NCT06586905 (Registered Sept 4, 2024)
Detecting the North Atlantic Right Whale Using Satellite Imagery
No full textThe North Atlantic right whale (Eubalaena glacialis) is critically endangered, with only about 370 individuals remaining. Modern conservation efforts rely on accurate knowledge of their location and movements, and, while established surveying methods such as aerial survey flights produce high-quality data for this purpose, they can be costly and are unable to cover large areas. Satellite imaging has been proposed as an additional tool to aid in the detection and monitoring of the whales, allowing for much broader coverage at a lower cost, though at a reduced accuracy. This thesis describes the first attempts at observing the North Atlantic right whale in satellite imagery, and the development of an automated detector model, including a new form of training data.
On April 24th, 2021, concurrent WorldView-3 satellite imagery and aerial photographs were acquired in Cape Cod Bay, Massachusetts. Ideal environmental conditions and an abundance of whales in the area resulted in 39 whale observations in the imagery, which were confirmed by the aerial survey. It was demonstrated that North Atlantic right whales were fairly easily visible in 30 cm and 15 cm satellite imagery, and that they were able to be identified on a species level due to visible markings unique to the right whale.
While right whales are often easily visible in such satellite imagery, visually identifying whales in a large number of images would be a very slow and tedious task. To develop an automated whale detector model, a large number of examples of right whales in satellite imagery are needed to allow the model to "understand" all the different ways a whale can look in such imagery However, at the time only the 39 observations mentioned above were available. Here, aerial photographs were modified to resemble satellite imagery using a deep learning approach called Neural Style Transfer (NST), in which the style of an existing satellite image of a right whale is transferred to the content of an aerial photograph. A unique set of satellite/aerial 'reference pairs' was developed, allowing for direct comparison between actual satellite imagery and the newly developed 'satellite-like' NST images using image similarity metrics. This demonstrated that the NST images were significantly more similar to satellite imagery than unmodified aerial photographs, and allowed for the immediate increase of examples of right whales in 'satellite imagery' from 39 to many thousands.
The NST images were directly compared to other types of training imagery, including unmodified aerial photographs, colour-normalized aerial photographs, and satellite imagery itself, by training a detector model using each of these training data types and comparing detection accuracies between them. It was found that models trained on NST simulated images slightly outperformed those trained on colour normalized photographs, while both significantly outperformed unmodified aerial photographs. Models trained on satellite imagery had an excellent precision but rather poor recall, likely due to the small amount of training data relative to the other datasets. These results indicate that performing some type of preprocessing modification to aerial photographs before training is highly desirable, though the trade-off between a slight increase in accuracy with NST and the significantly lower preprocessing time with colour normalized photographs will be a decision point for the end user
Enhancing Object Detection with Transformer-Based Adaptive Sensor Fusion
No full textAchieving reliable perception in dynamic environments while enabling real-time decision-making is critical for practical deployment in autonomous vehicles. The objective of this research is to enhance the accuracy, robustness, and computational efficiency of object detection systems for autonomous driving.
To address the need for efficient and low-latency, we first developed TransfuseNet, a lightweight LiDAR-camera fusion network specifically designed for 2D object detection. TransfuseNet optimizes computational efficiency by leveraging self-attention mechanisms for mid-level feature fusion and introducing a Multi-Convolutional Fusion (MCF) operator that prioritizes essential features. With its compact model architecture and reduced resource consumption, TransfuseNet achieves inference latency below 40ms, making it well-suited for real-time applications where rapid action is required. However, while TransfuseNet effectively balances accuracy and efficiency, it does not explicitly account for sensor reliability variations or provide mechanisms to adapt to degraded sensor inputs.
To overcome these limitations, we introduced ReliFusion, a reliability focused LiDAR-camera fusion framework for 3D object detection. ReliFusion was designed as a more advanced fusion model that integrates LiDAR and camera data for enhanced perception and dynamically adjusts sensor contributions based on real-time reliability assessments. Unlike conventional fusion strategies that assume equal reliability of all modalities, ReliFusion incorporates adaptive mechanisms to ensure robustness under sensor degradation, occlusions, and environmental challenges. It integrates a Spatio-Temporal Feature Aggregation (STFA) module to improve temporal consistency, a Reliability module based on Cross-Modality Contrastive Learning (CMCL) to quantify the trustworthiness of sensor inputs, and a Confidence-Weighted Mutual Cross-Attention (CW-MCA) module to refine fusion weights according to estimated reliability scores. This adaptive approach enables ReliFusion to maintain stable detection performance even in challenging real-world conditions.
Experimental evaluations on the KITTI and nuScenes datasets demonstrate that both TransfuseNet and ReliFusion achieve improved detection accuracy compared to existing fusion-based methods. While TransfuseNet provides an efficient solution for real-time 2D detection, ReliFusion advances multimodal 3D detection by addressing sensor degradation and incorporating dynamic reliability-driven fusion strategies. The findings of this research contribute to the design of sensor fusion-based object detection systems that enhance multimodal perception in autonomous vehicles by addressing key challenges such as sensor degradation, occlusions, and dynamic environmental conditions
Development of Novel Alginate-Based Bioinks for 3D Bioprinting of Cartilagenous Tissues
No full text3D bioprinting holds promise for a broad range of tissue engineering applications, showing particular potential in the development of engineered articular cartilage grafts for treating cartilage injuries. However, the development of bioinks with optimal rheological, mechanical, and biological properties remains a significant challenge. Alginate is a commonly used bioink base polymer and is often prepared by pre-crosslinking with Ca²⁺ prior to extrusion-based bioprinting to improve its rheological properties and ensure an acceptable degree of shape fidelity. This study examines the use of alternative divalent cations as pre-crosslinking agents by evaluating and optimizing printability, shape fidelity, and cell viability. Cations with high binding affinities to alginate are found to improve the printability of bioinks. In particular, Sr²⁺ pre-crosslinked bioinks are found to support high chondrocyte viability and enhance the bioink's shape retention, yet lower the shear force required for flow, when compared to Ca²⁺. Constructs made with Sr²⁺ and Ca²⁺ pre-crosslinked bioinks were printed and post-crosslinked with either Sr²⁺ or Ca²⁺ for further evaluation. The Sr²⁺ pre-crosslinked bioinks are found to yield constructs that exhibit decreased degradation rates and increased mechanical strength independently of the choice of post-crosslinking cation and can support the accumulation of cartilage extracellular matrix by encapsulated chondrocytes. Analysis of cation retention finds that the pre-crosslinker is substantially replaced by the post-crosslinking agent, indicating that the pre-crosslinking agent may impact the construct properties and cell responses primarily through changes in the hydrogel microstructure. Taken together, the results of this study indicate that the choice of pre-crosslinker for alginate bioinks holds promise as a simple method of improving upon bioink shape fidelity and tuning bioprinted tissue construct properties
Strengthening Evidence-Based Vaccine Communication Through Knowledge Mobilization: COVID-19 as an Exemplar
No full textPractical considerations for residential-managed alcohol programs: lessons from Ottawa Inner City Health
No full textAbstract Background Alcohol Use Disorder (AUD) is a leading contributor to global morbidity and mortality, disproportionately affecting people experiencing homelessness. Managed Alcohol Programs (MAPs) represent a harm reduction-based strategy for individuals with severe AUD and homelessness, providing controlled amounts of alcohol alongside comprehensive health and social supports. While evidence of MAP benefits continues to grow, important questions remain about how best to integrate social and medical care, and how to tailor services to align with participants’ goals, values, and broader social and structural contexts. Main body This commentary explores the operational strategies and clinical practices of the Ottawa Inner City Health (OICH) MAP, which has been running since 2001. We describe how the program is embedded within supportive housing and leverages an interdisciplinary team—including peer workers and an Indigenous healer—to deliver person-centered care. Key components include structured alcohol delivery tailored to individual needs, meal provision, social supports including life skills training, medication administration and comprehensive physical and mental health services. Clinical care is tailored to participants’ day-to-day circumstances, challenges, and goals in managing their AUD, with particular attention to hygiene and nutrition, proactive screening for health decline, and timely management of common health complications. The program operates through strong partnerships with community organizations, pharmacies and subspecialists, to enable integrated, coordinated care. Collaborative and trauma-informed approaches reduce reliance on emergency care and foster a sense of dignity, stability, and community. Conclusion MAPs have evolved from experimental interventions into internationally recognized harm reduction models. The OICH MAP demonstrates how the integration of housing, healthcare, and social supports can address the complex needs of individuals experiencing homelessness and severe AUD. However, challenges remain in scaling these models, refining screening protocols, and developing evidence-based policy frameworks. This commentary offers practical insights to inform the effective operation of MAPs and calls for continued research and dialogue to ensure they remain adaptable, sustainable, and aligned with the realities of the populations they serve
Towards Sustainable Construction: Innovations in Hemp-Lime Composite Production for Carbon-Negative Building Materials
No full textHemp-lime composites (HLC) have captured significant attention in the construction industry due to their sustainability, availability, and excellent hygrothermal performance. These bio-based insulating materials are produced by mixing a lime-based binder with hemp shivs and water. Despite their positive attributes, the properties and performance of the resulting composite depend on various parameters such as the ratio of constituent materials, manufacturing method, binder type, and shiv characteristics (e.g., particle size and purity). Therefore, its widespread adoption has been impeded by the notable variation in hemp-lime formulation and manufacturing methods, leading to inconsistent performance. In this research, new methods are proposed to enhance the homogeneity and uniformity of hemp-lime composites while minimizing their environmental impact to produce a carbon-negative insulating material with predictable and reproducible properties. The main objectives of this research are: 1) to improve the consistency and repeatability of hemp lime composites performance by reducing the hemp shiv particle size and introducing vibration as a compaction method; 2) to investigate the effect of impurities (e.g., hemp fibres) on the mechanical and hygrothermal properties of hemp lime composites; and 3) to investigate the effectiveness of producing hydrated lime from different CaCO₃ sources, including organic waste, using an electrochemical decarbonation process to formulate carbon-negative hemp lime composites for sustainable construction.
In accordance with the above objectives, modifications are proposed to the hemp shiv particle size, binder source, and placement method used to make hemp-lime composites. The experimental results demonstrate a marked improvement in hygrothermal properties compared to values reported in available literature, with thermal conductivity in the range of 0.053-0.060 W/m K and moisture buffering values between 2.21 to 2.53 g/m² RH, as well as a very low coefficient of variation of dry density between 0.37-0.73%. The results also demonstrate a noteworthy reduction in the directional difference of thermal conductivity, reduced to less than 3% when employing fine hemp fragments. Furthermore, using finer particles reduces the amount of binder needed which significantly lowers the carbon footprint when using conventional hydrated lime produced through calcination. An alternative process is also introduced in which hydrated lime is produced through electrolysis, which facilitates direct carbon capture. This process is evaluated using three different limestone sources and two organic sources (mussel and eggshells) as feedstock. The precipitated materials are mainly comprised of Ca(OH)₂ with a suitable chemical composition and particle size distribution required for Type N hydrated lime. The environmental assessment of wall assemblies revealed that hemp-lime composites incorporating electrochemically decarbonated hydrated lime (ED-HLC) offer comparable thermal and moisture performance to conventional binders while achieving superior mechanical strength. Life-cycle analysis demonstrated a markedly lower embodied footprint and a carbon-negative balance, reaching -23.1 kg CO₂ eq. when carbon sequestration was included. Compared to conventional lime-based hemp-lime composites and glass wool insulation, the ED-HLC wall showed 15.7% and 457% lower global warming potentials, respectively, confirming its strong potential as a carbon-conscious insulation solution
Survey of Ontario Pharmacists' Knowledge, Resources and Barriers in Providing Perinatal Care
No full textBackground: Medication use during the perinatal period is challenging, as ethical concerns often exclude these patients from trials. With evolving perinatal care needs, clinicians continue to rely on limited evidence. Few Ontario pharmacists specialize in this area, highlighting gaps in knowledge, confidence and training. Objectives: To (1) explore Ontario pharmacists’ experience, knowledge, and practices in providing care during preconception, pregnancy, and breastfeeding; (2) identify the resources pharmacists use for reliable information on perinatal medication use; and (3) determine the barriers to pharmacy care for conceiving, pregnant, and breastfeeding patients.
Methods: An electronic survey in English and French was sent to active Ontario pharmacists, collecting demographics, experience, self-assessed knowledge, resources, and barriers in caring for conceiving, pregnant, and breastfeeding patients. Data were analyzed anonymously using descriptive statistics in Excel.
Results: While 92% completed training in the past year, 60% reported none in perinatal health, revealing educational gaps. Respondents reported infrequent patient counseling and lower confidence in preconception care. Main barriers included difficulty interpreting data (51%), limited resource awareness (39%), time constraints (37%), and lack of education (33%).
Conclusions: Findings show significant gaps in pharmacists’ preparedness for perinatal care. Despite general education and continuing education, most lacked specific training and confidence in perinatal care. Enhanced perinatal medication training should be included in pharmacy education, with clear guidance on using specialized references to support consistent, evidence-based counseling and improve care quality
Informing Policy: Does Television Food and Beverage Advertising Not Specifically Targeting Children Influence Their Behaviour?
No full textThis research examined the frequency of child-targeted versus non-child-targeted food and beverage advertising on children's television, comparing two distinct food marketing policy environments in Canada (Quebec and Ontario). It also investigated children's perceptions of both types of advertising and identified the features that elicit positive attitudes and influence behavioural responses. Finally, the study explored whether parental exposure to unhealthy food marketing is associated with children's consumption and requests for such products, and whether these behavioural outcomes differ across countries. A mixed-methods design addressed the research questions. Study 1 involved a cross-sectional analysis of advertisements (ads) on popular children's television channels in Ontario and Quebec, with descriptive statistics tabulated for the frequency of ads and marketing techniques classified as child-targeted and non-child-targeted. Study 2 comprised open-ended online interviews with 17 children recruited through convenience sampling, during which participants viewed four ads: one child-targeted and one non-child-targeted ad for both healthy (plain milk) and unhealthy (chocolate) foods. A thematic analysis was conducted. Study 3 consisted of a secondary analysis of the 2018 International Food Policy Study, including 5,764 parents of children under 18 from Australia, Canada, Mexico, the United Kingdom, or the United States. Binary logistic regressions assessed associations between parental exposure to unhealthy food marketing and children's purchase request and actual purchasing outcomes. Results indicated that food and beverage advertising represented a higher proportion of total advertising in Quebec, with Ontario showing a higher frequency of child-targeted ads and Quebec a higher frequency of non-child-targeted ads, including alcohol ads. In Study 2, pre-existing familiarity with products was more relevant than marketing techniques. No notable differences were observed between child-targeted and non-child-targeted ads. Study 3 found that greater parental exposure to unhealthy food marketing was associated with higher purchase request and purchase outcomes. These findings highlight critical gaps in current regulatory frameworks. Food marketing is a complex and persistent public health challenge that requires coordinated, multi-level policy responses. Effective policies must go beyond narrow definitions and adopt broader, exposure-based strategies. This thesis contributes evidence to support stronger, more comprehensive regulations that reflect the diverse ways in which marketing reaches and influences children and families
Predicting child and adolescent mental health emergency department revisits: a machine-learning approach compared to a clinician-derived baseline
No full textAbstract Background Predicting child and youth mental health (CYMH) emergency department (ED) revisits (RVs) is critical for improving patient outcomes and optimizing use of resources. Previous CYMH ED RV studies have used statistical methods with research cohorts and produced varying results. Our aims were to develop a predictive algorithm incorporating machine learning (ML) with electronic health records (EHR) and validate it against a clinician-driven algorithm in a proof of concept project. Methods Data were retrospectively collected from a tertiary care pediatric hospital’s EHR from November 2017–November 2023, yielding 12,700 ED encounters from 8,696 patients, 8–18 years of age. The feature set comprised patient demographics, visit-level variables, laboratory results, procedure codes, and medication records. A mapping of 230 International Classification of Diseases (ICD)-10 codes into 28 Diagnostic and Statistical Manual (DSM)-5 categories was performed and a logistic regression (LR) ML model developed. Both tasks used clinical expert input. Seven clinical experts then independently assigned weights to 191 variables using a custom-designed application to create a structured clinician-weighted baseline for comparison to the ML algorithm. Both models were evaluated using AUROC and F1 score as primary metrics with precision and recall as secondary. LR coefficients and odds ratios were the primary interpretability outputs, while SHapley Additive exPlanations (SHAP) were used for supplementary visualization across four age strata. Results The LR machine learning model achieved an AUROC of 0.78, outperforming the structured clinician-weighted baseline (AUROC range: 0.54–0.64) Detailed analysis revealed that predictors such as past ED RV count, psychotherapeutic medication history, substance use history, and prior outpatient MH visits were consistently influential. Conclusions This proof of concept project demonstrates that ML can provide complementary, clinically interpretable predictions of CYMH ED RV. Alignment between model-derived predictors and clinician-weighted features supports interpretability and lays a foundation for further development. Future steps include enhancing sensitivity, expanding feature sets, and conducting prospective silent-mode validation to refine performance. Clinical trial registration Not applicable