Canadian Institute Research Articles

Referred molecular testing as a barrier to optimal treatment decision making in metastatic non-small cell lung cancer: Experience at a tertiary academic institution in Canada.

Molecular testing is critical to guiding treatment approaches in patients with metastatic non-small cell lung cancer (mNSCLC), with testing delays adversely impacting the timeliness of treatment decisions. Here, we aimed to evaluate the time from initial mNSCLC diagnosis to treatment decision (TTD) following implementation of in-house EGFR, ALK, and PD-L1 testing at our institution. We conducted a retrospective chart review of 165 patients (send-out testing, n = 92; in-house testing, n = 73) with newly diagnosed mNSCLC treated at our institution. Data were compared during the send-out (March 2017-May 2019) and in-house (July 2019-March 2021) testing periods. We performed a detailed workflow analysis to provide insight on the pre-analytic, analytic, and post-analytic intervals that constituted the total TTD. TTD was significantly shorter with in-house testing (10 days vs. 18 days, p < 0.0001), driven largely by decreased internal handling and specimen transit times (2 days vs. 3 days, p < 0.0001) and laboratory turnaround times (TAT, 3 days vs. 8 days, p < 0.0001), with 96% of in-house cases meeting the international guideline of a ≤ 10-day intra-laboratory TAT (vs. 74% send-out, p < 0.001). Eighty-eight percent of patients with in-house testing had results available at their first oncology consultation (vs. 52% send-out, p < 0.0001), and all patients with in-house testing had results available at the time of treatment decision (vs. 86% send-out, p = 0.57). Our results demonstrate the advantages of in-house biomarker testing for mNSCLC at a tertiary oncology center. Incorporation of in-house testing may reduce barriers to offering personalized medicine by improving the time to optimal systemic therapy decision.

Exploring the Quality of Feedback in Entrustable Professional Activity Narratives Across 24 Residency Training Programs.

Background Competency-based medical education (CBME) has been implemented in many residency training programs across Canada. A key component of CBME is documentation of frequent low-stakes workplace-based assessments to track trainee progression over time. Critically, the quality of narrative feedback is imperative for trainees to accumulate a body of evidence of their progress. Suboptimal narrative feedback will challenge accurate decision-making, such as promotion to the next stage of training. Objective To explore the quality of documented feedback provided on workplace-based assessments by examining and scoring narrative comments using a published quality scoring framework. Methods We employed a retrospective cohort secondary analysis of existing data using a sample of 25% of entrustable professional activity (EPA) observations from trainee portfolios from 24 programs in one institution in Canada from July 2019 to June 2020. Statistical analyses explore the variance of scores between programs (Kruskal-Wallis rank sum test) and potential associations between program size, CBME launch year, and medical versus surgical specialties (Spearman's rho). Results Mean quality scores of 5681 narrative comments ranged from 2.0±1.2 to 3.4±1.4 out of 5 across programs. A significant and moderate difference in the quality of feedback across programs was identified (χ2=321.38, P<.001, ε2=0.06). Smaller programs and those with an earlier launch year performed better (P<.001). No significant difference was found in quality score when comparing surgical/procedural and medical programs that transitioned to CBME in this institution (P=.65). Conclusions This study illustrates the complexity of examining the quality of narrative comments provided to trainees through EPA assessments.

Unnecessary interventions for the management of hip osteoarthritis: a population-based cohort study.

Patients aged 40-60 years who require total hip arthroplasty (THA) often first receive unindicated hip arthroscopy or magnetic resonance imaging (MRI). Our objective was to identify potentially inappropriate resource utilization before THA, specifically reporting on the proportion of patients aged 40-60 years who underwent hip arthroscopy or MRI in the year before THA. We conducted a retrospective, population-based study at the provincial level. We retrieved data from the Canadian Institute for Health Information (CIHI). We included all Ontario residents who underwent an elective, primary THA for osteoarthritis between Apr. 1, 2004, and Mar. 31, 2016. We identified the rates and timing of patients who underwent an MRI or hip arthroscopy before their index THA. The percentage of patients who underwent an MRI before THA increased significantly over the study period, from 8.7% in 2004 to 23.8% in 2015. There was also a significant but variable trend in the percentage of patients who underwent a hip arthroscopy before THA. Our results demonstrate a high, gradually increasing proportion of patients who received a hip MRI and a low but increasing proportion of patients who received hip arthroscopy in close proximity to THA. Multidisciplinary collaboration may improve knowledge translation and help reduce the rate of clinically unnecessary diagnostic and therapeutic interventions in this population of patients who require THA.

A randomized trial evaluating the utility of non-targeted biopsies for colorectal neoplasia detection in adults with inflammatory bowel disease: a pilot study protocol

BackgroundPersons with inflammatory bowel diseases are at increased risk of developing colorectal cancer and require frequent colonoscopy surveillance. Guidelines recommend taking 30 to 40 non-targeted biopsies throughout the colorectum to detect “invisible” neoplasia in this setting, despite a lack of evidence supporting this practice. We sought to assess the utility of this practice through a randomized controlled trial. We first propose an internal pilot study to assess recruitment potential, protocol adherence and data capture to guide the full trial.MethodsWe have designed a multi-centre, parallel-group, non-inferiority randomized controlled trial to test the utility of non-targeted biopsies as an adjunct to colonoscopy surveillance for neoplasia detection in persons with inflammatory bowel disease involving the colorectum in routine clinical practice. Participants are randomized 1:1, stratified by study site, to either standard of care high-definition white-light colonoscopy with 32 to 40 non-targeted biopsies of non-neoplastic-appearing mucosa along with a sampling of abnormal-appearing mucosa (control group) or modified colonoscopy with targeted sampling alone (intervention group). The primary outcome for the full trial will be the proportion of persons with ≥ 1 neoplastic focus detected during colonoscopy. For the pilot phase, we will assess the feasibility of recruiting a minimum of 15% of the estimated sample size within 1 year, under identical conditions as the full trial, while maintaining ≥ 90–95% rate of protocol adherence and data capture. These participants will contribute data to the full trial. The trial is being conducted at 12 centres across Canada, with a total sample size of 1952 persons.DiscussionsThe trial protocol has been approved by the ethics committees of all participating sites, and the pilot study has received funding through the Canadian Institutes of Health Research (PJT 159607). If feasibility metrics are met during the pilot phase, we will complete the full trial. The trial outcomes will contribute to update the practice guidelines in this area.Trial registrationClinicalTrials.gov, NCT04067778.

Self-Supervised Self-Organizing Clustering Network: A Novel Unsupervised Representation Learning Method.

Deep learning-based clustering methods usually regard feature extraction and feature clustering as two independent steps. In this way, the features of all images need to be extracted before feature clustering, which consumes a lot of calculation. Inspired by the self-organizing map network, a self-supervised self-organizing clustering network ( [Formula: see text]OCNet) is proposed to jointly learn feature extraction and feature clustering, thus realizing a single-stage clustering method. In order to achieve joint learning, we propose a self-organizing clustering header (SOCH), which takes the weight of the self-organizing layer as the cluster centers, and the output of the self-organizing layer as the similarities between the feature and the cluster centers. In order to optimize our network, we first convert the similarities into probabilities which represents a soft cluster assignment, and then we obtain a target for self-supervised learning by transforming the soft cluster assignment into a hard cluster assignment, and finally we jointly optimize backbone and SOCH. By setting different feature dimensions, a Multilayer SOCHs strategy is further proposed by cascading SOCHs. This strategy achieves clustering features in multiple clustering spaces. [Formula: see text]OCNet is evaluated on widely used image classification benchmarks such as Canadian Institute For Advanced Research (CIFAR)-10, CIFAR-100, Self-Taught Learning (STL)-10, and Tiny ImageNet. Experimental results show that our method significant improvement over other related methods. The visualization of features and images shows that our method can achieve good clustering results.

Skin Antisepsis before Surgical Fixation of Extremity Fractures

BackgroundStudies evaluating surgical-site infection have had conflicting results with respect to the use of alcohol solutions containing iodine povacrylex or chlorhexidine gluconate as skin antisepsis before surgery to repair a fractured limb (i.e., an extremity fracture).MethodsIn a cluster-randomized, crossover trial at 25 hospitals in the United States and Canada, we randomly assigned hospitals to use a solution of 0.7% iodine povacrylex in 74% isopropyl alcohol (iodine group) or 2% chlorhexidine gluconate in 70% isopropyl alcohol (chlorhexidine group) as preoperative antisepsis for surgical procedures to repair extremity fractures. Every 2 months, the hospitals alternated interventions. Separate populations of patients with either open or closed fractures were enrolled and included in the analysis. The primary outcome was surgical-site infection, which included superficial incisional infection within 30 days or deep incisional or organ-space infection within 90 days. The secondary outcome was unplanned reoperation for fracture-healing complications.ResultsA total of 6785 patients with a closed fracture and 1700 patients with an open fracture were included in the trial. In the closed-fracture population, surgical-site infection occurred in 77 patients (2.4%) in the iodine group and in 108 patients (3.3%) in the chlorhexidine group (odds ratio, 0.74; 95% confidence interval [CI], 0.55 to 1.00; P=0.049). In the open-fracture population, surgical-site infection occurred in 54 patients (6.5%) in the iodine group and in 60 patients (7.3%) in the chlorhexidine group (odd ratio, 0.86; 95% CI, 0.58 to 1.27; P=0.45). The frequencies of unplanned reoperation, 1-year outcomes, and serious adverse events were similar in the two groups.ConclusionsAmong patients with closed extremity fractures, skin antisepsis with iodine povacrylex in alcohol resulted in fewer surgical-site infections than antisepsis with chlorhexidine gluconate in alcohol. In patients with open fractures, the results were similar in the two groups. (Funded by the Patient-Centered Outcomes Research Institute and the Canadian Institutes of Health Research; PREPARE ClinicalTrials.gov number, NCT03523962.)

Navigating the Paradigm Shift in Chronic Lymphocytic Leukemia Treatment from Chemotherapy to Targeted Therapies

The shift from traditional chemotherapy to more targeted therapies has been a landmark change in chronic lymphocytic leukemia (CLL) treatment. This transformation has implications for treatment efficacy, tolerability, and patient quality of life, along with implications for the Canadian oncology community at large, which must rapidly adapt to these advancements. This rapid development underscores the importance of continued responsiveness in medical practice, including more collaborative work with Canadian institutions and provincial cancer care to deliver these transformative therapies to patients. This review aims to offer practical guidance, from a Canadian perspective, for clinicians in treatment selection in the era of targeted therapies.

‘Safer supply’ in BC tied to increase in poisonings, but not deaths

A cohort study that looked at Canadian province British Columbia's Safer Opioid Supply policy, instituted two years ago, found that there was a moderate increase in the number of individuals with at least one opioid prescription, a large increase in the number of opioid prescriptions dispensed, and a substantial increase in hospitalizations related to opioid poisoning. According to the study, “British Columbia's Safer Opioid Supply Policy and Opioid Outcomes” published online in JAMA Internal Medicine on Jan. 16, there were no statistically significant changes in deaths from opioid overdoses and no significant change in the number of prescribers. Rather, those prescribers who were there before prescribed a significantly greater amount of opioids. And the opioid‐related poisoning hospitalization rate increased by 3.2 per 100,000 population. The study, by Hai V. Nguyen, Ph.D. and colleagues, was funded by the Canadian Institutes of Health Research. The study used quarterly data from 2016 to 2022 from British Columbia, where the Safer Opioid Supply policy was implemented, then compared that data to Canada's Manitoba and Saskatchewan provinces where the policy was not implemented. The main outcomes were rates of prescriptions, claimants, and prescribers of opioids targeted by the Safer Opioid Supply policy (hydromorphone, morphine, oxycodone, and fentanyl); opioid‐related poisoning hospitalizations; and deaths from apparent opioid toxicity. The researchers concluded that while the Safer Opioid Supply policy was associated with higher rates of safer (prescription) supply, it was also associated with a significant increase in opioid‐related poisoning hospitalizations. “These findings will help inform ongoing debates about this policy not only in British Columbia, but also in other jurisdictions that are contemplating it,” the researchers concluded.

Dimensions of Local Authority: Mapping Local Political Institutions in Canada's Cities

Abstract Our knowledge of the institutional features of local government in Canadian cities is surprisingly fragmentary. The academic literature has long identified dominant tendencies in Canadian local institutions, but systematic empirical data has been missing. In this article, we address this gap in knowledge in two ways. We introduce the Canadian Municipal Attributes Portal (CMAP), a new open-access database that contains information on dozens of institutional features of local government for nearly 100 of the most populous municipalities in Canada. We then propose a new multidimensional index of authority concentration, which is designed to capture variation in the local structure of decision-making authority in a systematic and nuanced manner. We apply this index to a systematic pan-Canadian subsample of 65 CMAP municipalities. The result is a rich portrait of institutional variety, one that both corroborates and substantially extends our current understanding of the shape of municipal institutions in Canadian cities.

Evaluating deep learning variants for cyber-attacks detection and multi-class classification in IoT networks.

The Internet of Things (IoT), considered an intriguing technology with substantial potential for tackling many societal concerns, has been developing into a significant component of the future. The foundation of IoT is the capacity to manipulate and track material objects over the Internet. The IoT network infrastructure is more vulnerable to attackers/hackers as additional features are accessible online. The complexity of cyberattacks has grown to pose a bigger threat to public and private sector organizations. They undermine Internet businesses, tarnish company branding, and restrict access to data and amenities. Enterprises and academics are contemplating using machine learning (ML) and deep learning (DL) for cyberattack avoidance because ML and DL show immense potential in several domains. Several DL teachings are implemented to extract various patterns from many annotated datasets. DL can be a helpful tool for detecting cyberattacks. Early network data segregation and detection thus become more essential than ever for mitigating cyberattacks. Numerous deep-learning model variants, including deep neural networks (DNNs), convolutional neural networks (CNNs), and recurrent neural networks (RNNs), are implemented in the study to detect cyberattacks on an assortment of network traffic streams. The Canadian Institute for Cybersecurity's CICDIoT2023 dataset is utilized to test the efficacy of the proposed approach. The proposed method includes data preprocessing, robust scalar and label encoding techniques for categorical variables, and model prediction using deep learning models. The experimental results demonstrate that the RNN model achieved the highest accuracy of 96.56%. The test results indicate that the proposed approach is efficient compared to other methods for identifying cyberattacks in a realistic IoT environment.

Gabapentinoids and Risk for Severe Exacerbation in Chronic Obstructive Pulmonary Disease : A Population-Based Cohort Study.

North American and European health agencies recently warned of severe breathing problems associated with gabapentinoids, including in patients with chronic obstructive pulmonary disease (COPD), although supporting evidence is limited. To assess whether gabapentinoid use is associated with severe exacerbation in patients with COPD. Time-conditional propensity score-matched, new-user cohort study. Health insurance databases from the Régie de l'assurance maladie du Québec in Canada. Within a base cohort of patients with COPD between 1994 and 2015, patients initiating gabapentinoid therapy with an indication (epilepsy, neuropathic pain, or other chronic pain) were matched 1:1 with nonusers on COPD duration, indication for gabapentinoids, age, sex, calendar year, and time-conditional propensity score. The primary outcome was severe COPD exacerbation requiring hospitalization. Hazard ratios (HRs) associated with gabapentinoid use were estimated in subcohorts according to gabapentinoid indication and in the overall cohort. The cohort included 356 gabapentinoid users with epilepsy, 9411 with neuropathic pain, and 3737 with other chronic pain, matched 1:1 to nonusers. Compared with nonuse, gabapentinoid use was associated with increased risk for severe COPD exacerbation across the indications of epilepsy (HR, 1.58 [95% CI, 1.08 to 2.30]), neuropathic pain (HR, 1.35 [CI, 1.24 to 1.48]), and other chronic pain (HR, 1.49 [CI, 1.27 to 1.73]) and overall (HR, 1.39 [CI, 1.29 to 1.50]). Residual confounding, including from lack of smoking information. In patients with COPD, gabapentinoid use was associated with increased risk for severe exacerbation. This study supports the warnings from regulatory agencies and highlights the importance of considering this potential risk when prescribing gabapentin and pregabalin to patients with COPD. Canadian Institutes of Health Research and Canadian Lung Association.

Gender and Racial Diversity Among Obstetrics and Gynecology Departments and Gynecologic Oncology Divisions in Canada: Are We There Yet?

ObjectivesGender and racial diversity in academic Canadian departments of obstetrics and gynecology (OBGYN) have not been previously described. We examined gender representation in leadership in academic OBGYN departments and gynecologic oncology (GO) divisions, and determined factors predictive of leadership and promotion including racialized status. MethodsThis cross-sectional study of Canadian residency-affiliated academic OBGYN departments queried institutional websites in January 2021 to compile a list of academic faculty. Subjective gender was assessed using photographs and pronouns, and racialized status was determined using photographs. Logistic regression analyses determined predictive factors for leadership roles. Fassiotto et al. rank equity indices (REI) and Hofler et al. representation ratios were calculated. ResultsWithin 16 Canadian institutions there were 354 (33.6%) men and 699 (66.4%) women, with 18.3% racialized faculty. Men were more likely to reach full professorship (P < 0.00001) and leadership positions of department chair, vice-chair or division head (P = 0.01). Representation ratios for women in OBGYN were <1 for all administrative leadership positions, and pairwise comparisons of the probability of promotion for women OBGYNs using REI reveal significant disparities between senior and junior administrative leadership and professorial ranks. Racialized physicians were less likely to have attained full professorship (P = 0.002). Ninety-seven academic GOs were identified: 68 (70.1%) were women, 17 (17.5%) racialized. Seven GO divisions (44%) had no racialized members. On multivariate analysis, only year of completion of fellowship was predictive of leadership. ConclusionIn academic Canadian OBGYN departments women are underrepresented in leadership and full professor positions. Racialized faculty are underrepresented in full professorship.

Assessment of h-index and associated demographic and academic parameters for academic hematologists in Canada.

The h-index measures researchers' productivity by assessing simultaneously the number of publications and citations. We aimed to assess the factors that could influence h-index for hematologists practicing in academic institutions in Canada. We identified universities with a hematology residency training programs/fellowships using the Canadian Resident Matching Service (CaRMS) website. We obtained the listing of faculty, sex, and academic ranks by consulting faculty directories or by contacting respective departments/universities, when directories were unavailable or incomplete. For each faculty member, we obtained years since Royal College of Physicians' and Surgeons of Canada certification or equivalent, receipt of Canadian Institute of Health Research (CIHR) grants within the last 5 years, attainment of graduate degrees (M.Sc., Ph.D., other), and the h-index. The data included information collected from 372 individuals (171 females) across Canada (Atlantic Provinces: 13; Quebec: 89; Ontario: 182; Prairie Provinces: 59; British Columbia: 29). Univariate analysis showed that male sex, practicing in British Columbia, longer duration since specialty certification, completion of an M.Sc. or a Ph.D. degree, attaining a higher academic rank and receiving CIHR funding were associated with higher h-index. The results of the univariate analysis were concordant with the multivariate analysis, except that practicing in Ontario was also associated with higher h-index. This study provides details on the h-index curve/parameters for academic productivity of hematologists in Canada. Importantly, based on multivariate analysis, higher h-index was associated with male sex, location of practice, years since certification, attainment of M.Sc. or Ph.D. degrees, academic rank, and recent CIHR funding.

Overview and Exploratory Analyses of CICIDS2017 Intrusion Detection Dataset

Intrusion detection systems are used to detect attacks in a network. Machine learning (ML) approaches have been widely used to build such Intrusion Detection Systems (IDSs) because they are more accurate when built from a very large and representative dataset. In recent times, one of the benchmark datasets that are used to build ML-based Intrusion Detection models is CICIDS2017 dataset. The dataset is contained in eight groups and was collected from the Canadian Institute on Cyber Security dataset repository. The dataset is available in both PCAP and netflow formats. This study used the netflow records in the CIDIDS2017 dataset as they are found to contain newer attacks,very large and are found useful for traffic analysis. Exploratory Data Analysis (EDA) techniques were used to reveal various characteristics of the dataset. The general objective is to provide more insights on the nature, structure and issues with the dataset so as to identify the best ways for using it to achieve improved ML-based IDS models. Furthermore, some of the open problems that can arise from the use of the dataset in any machine learning-based Intrusion Detection systems are highlighted and possible solutions are briefly discussed. The EDA techniques used revealed important relationships among input variables and the target class. The study concluded that the EDA can better influence the decision of future IDS researches that use the dataset. Thus, improved machine learning-based intrusion detection systems can be built from the dataset once it is well understood and pre-processed.

Research Ethics Oversight for Multi-jurisdictional Clinical Trials in Canada: A Historical Perspective to Inform Future Direction

As a condition of funding from the Canadian Tri-Agencies, researchers and institutions are expected to adhere to the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (TCPS) and research projects must obtain approval from a Research Ethics Board (REB). Yet there are a limited number of frameworks and standards to guide the conduct and decisions of REBs in Canada and these are most often voluntary. Delays and increased costs resulting from the lack of a common or coordinated approach to REB review for multi-jurisdictional clinical trials is a long-standing issue in the Canadian research environment. Formal reviews and recommendations as early as 2006 call for accreditation or qualification for REBs, use of common forms and templates, and federal leadership to harmonize processes.

A qualitative system review: To lead public policy advancement for workplace psychological violence injury

This article presents a qualitative system analysis informing public policy advancement for workplace psychological violence injury. It is a unique look into how a jurisdiction responds to the injury caused by workplace psychological violence (workplace bullying, harassment, and discrimination campaigns). First ever provincial legislation addressing this harm was passed into law. This study found that the legislation has yet to be implemented into the medical response systems dealing with this workplace injury. Severe hardships and consequences are experienced by workers in attempting recovery. Some never get there. The Canadian Institute for Workplace Bullying Resources, a specialist with this injury, received this study's recommendations for system change to assist in closing gaps and reduce the harm experienced by workers in Alberta, Canada. This study found some causal factors to these system barriers to be common globally. Methods to map and lead advocacy system change through qualitative research for this workplace violence and caused injury are provided herein. This is a call to researchers and professionals globally who deal with both workplace psychological violence and injury, to use and build on this study's methods, tools, and outcomes to lead further needed system change.

The Role of Social Support in Concussion Rehabilitation: A Prospective Mixed Methods Study of Canadian University Athletes’ Return to Sport

Context: Social support is an important consideration during athletes’ rehabilitation following a concussion. Yet, its influence during the return-to-sport (RTS) strategy has been underexplored. Design: Concussed athletes’ experiences with social support during RTS were explored prospectively using an explanatory sequential mixed methods design. Methods: University rugby, basketball, and track and field athletes (N = 8) from a single Canadian institution, 19–23 years of age (mean = 20.63, SD = 1.51), completed concentric circles maps, the Perceived Available Support in Sport Questionnaire, and 2 semistructured interviews during their RTS. We analyzed concentric circles maps and questionnaire data descriptively and interview data using codebook thematic analysis. Results: Athletes identified 16 different social support agents. The 5 most important agents during RTS were athletic therapists, student therapists, head coaches, significant others, and teammates/friends. Significant others provided the most emotional (mean = 2.82, SD = 1.85) and esteem (mean = 2.63, SD = 1.85) support, and the athletic therapist (mean = 1.82, SD = 1.35) and head coaches (mean = 0.93, SD = 1.19) provided the most informational and tangible support. We generated 4 themes from athletes’ interviews: social support behaviors, contextual factors, concussion and RTS factors, and psychological readiness factors. Conclusions: The number of social support agents present, frequency of support types, and demonstration of support behaviors decreased across the concussion RTS strategy. Contextual (ie, toughness), concussion (ie, visibility), and RTS (ie, prognosis uncertainty) factors influenced athletes’ perceptions of social support agents’ support behaviors. Results add to our limited understanding of athletes’ social support during RTS following a concussion and suggest exploring the integration of behavior-specific (eg, checking in, providing reassurance, and demonstrating compassion) social support strategies to promote concussion rehabilitation and readiness to RTS.

Disease-modifying drugs, multiple sclerosis and infection-related healthcare use in British Columbia, Canada: a population-based study

Much remains unknown surrounding the disease-modifying drugs (DMDs) used to treat multiple sclerosis and infection-related healthcare use in the 'real-world' setting. We examined if DMD exposure was associated with altered infection-related healthcare use. We assessed if DMD (versus no) exposure was associated with altered infection-related hospitalizations, physician claims, and prescriptions filled in British Columbia, Canada (1996-2017). Healthcare use was assessed using negative binomial and proportional means regression models, reported as sex-/age-/comorbidity-/calendar year-/socioeconomic-adjusted rate and hazard ratios [aRR, aHR], with 95% confidence intervals [CIs]). We identified 19,360 multiple sclerosis cases (13,940/19,360; 72.0% women; mean age at study start=44.5 standard deviation, SD=13.3; mean follow-up=11.7 [SD=7.3] years). Relative to unexposed periods, exposure to any DMD was associated with a lower infection-related rate of physician claims (aRR=0.88; 95% CI:0.85-0.92) and hazard of hospitalization (aHR=0.64; 95% CI:0.56-0.73), and a higher rate of infection-related prescriptions (aRR=1.14; 95% CI:1.08-1.20). Exposure to any injectable or oral DMD was associated with a lower infection-related rate of physician claims (injectable aRR=0.88; 95% CI:0.84-0.92, oral aRR=0.83; 95% CI:0.77-0.90) and hazard of hospitalization (injectable aHR=0.65; 95% CI:0.56-0.75, oral aHR=0.54; 95% CI:0.38-0.77), whereas intravenous DMD exposure was not (aRR=0.99; 95% CI:0.86-1.14, aHR=0.73; 95% CI:0.49-1.09). Exposure to any injectable or intravenous DMD was associated with a higher rate of infection-related prescriptions (injectable aRR=1.15; 95% CI:1.08-1.22, intravenous=1.34; 95% CI:1.15-1.56), whereas oral DMDs were not (aRR=0.98; 95% CI:0.91-1.05). DMD exposure for the treatment of MS was associated with differences in infection-related healthcare use. While infection-related hospitalizations and physician visits were lower, prescription fills were higher. How these differences in infection-related healthcare use affect outcomes in persons with multiple sclerosis warrants consideration. Canadian Institutes of Health Research (CIHR); German Research Foundation (DFG).

FPGA based Flexible Implementation of Light Weight Inference on Deep Convolutional Neural Networks

Standard Convolution (StdConv) is the main technique used in the state of the art Deep Convolutional Neural Networks (DCNNs). Fewer computations are achieved if Depthwise Separable Convolution technique (SepConv) is used as an alternative. A crucial issue in many applications like smart cameras and autonomous vehicles where low latency is essential stems from deploying a lightweight and low cost inference models. An acceptable accuracy should be kept with tolerable computations and memory access load. A flexible architecture for different DCNN convolution types and models is proposed. The flexibility comes from the sharing of one memory access unit with different types of layers regardless of the selected kernel size, by multiplying each weight vector by local operators with variant aperture. Moreover, one depthwise computation unit can be used for both standard and pointwise layers. The learnable parameters are quantized to 8-bits fixed point representation and that gives very limited reduction of accuracy and a considerable reduction of the Field-Programmable Gate Array (FPGA) resources. To reduce processing time, inter layer parallel computations are performed. The experiment is conducted by using grey scale ORL database with shallow Convolutional Neural Network (CNN) and the colored Canadian Institute for Advanced Research 10 classes (CIFAR-10) database with DCNN, and a comparable accuracies of 93% and 85.7% are achieved respectively using very low cost of Spartan 3E and moderate cost of zynq FPGA platforms

Healthcare outcomes and dispositions in persons with obesity within emergency departments in Ontario, Canada: A cross-sectional analysis of the National Ambulatory Care Reporting System (NACRS), 2018-2022.

The experience of persons with obesity (PwO) in the Canadian healthcare setting has not been widely studied. The objective of this study was to assess care in PwO in emergency departments in Ontario, Canada. This secondary analysis made use of 2018-2022 Canadian Institute for Health Information's National Ambulatory Care Reporting System. The sample consisted of 4547 individuals with an obesity diagnosis, and 4547 controls who were matched for sex, age, and main diagnosis. Ordinal logistic and multiple linear regression analyses were used to assess triage scores, wait times, and length of stay. PwO had 4.8 minutes longer wait time for a physician initial assessment (p<0.01), 3.56 hours longer length of stay in the emergency department (p<0.0001), and 55% greater odds (OR = 1.55, 95% CI: 1.43-1.68) of having a less urgent triage score compared to controls matched for main diagnosis. When further matched for triage score, PwO experienced over three hours longer length of stay for triage level 2 (emergent, p<0.01), five hours longer for triage level 3 (urgent, p<0.01), and nearly two hours longer for triage level 4 (less urgent, p<0.05) cases. PwO were rated as less urgent and experienced longer wait times and length of stay, compared to controls matched by sex, age, and main diagnosis. Additional research is needed to confirm the consistency of these findings in other provinces/territories, and to examine clinical outcomes, and the underlying reasons for differences.