Generating Synthetic Images & Data to Improve Object Detection in CCTV Footage from Public Transport

Passenger behaviour on public transport has become a source of great interest in the wake of the COVID-19 pandemic. Operators are interested in employing new methods to monitor vehicle utilisation and passenger behaviour. One way to do this is through the use of Machine Learning, using the CCTV footage that is already being captured from the vehicles. However, one of the limitations of Machine Learning is that it requires large amounts of annotated training data, which is not always available. In this poster, we present a technique that uses 3D models to generate synthetic training images/data and discuss the effect that training with the synthetic data had on the Machine Learning models when applied to real-world CCTV footage. © 2022 IEEE.

On the Complexity of Object Detection on Real-world Public Transportation Images for Social Distancing Measurement

Social distancing in public spaces has become an essential aspect in helping to reduce the impact of the COVID-19 pandemic. Exploiting recent advances in machine learning, there have been many studies in the literature implementing social distancing via object detection through the use of surveillance cameras in public spaces. However, there has been no study of social distance measurement on public transport to date. The public transport setting has some unique challenges, including low-resolution images and physical camera locations that can lead to the partial occlusion of passengers, making it challenging to perform accurate detection. Thus, this paper investigates the challenges of performing accurate social distance measurements on public transportation. We benchmark several state-of-the-art object detection algorithms using real-world footage taken from the London Underground and bus network. The work highlights the complexity of performing social distancing measurements on images from current public transportation onboard cameras. Further, exploiting domain knowledge of expected passenger behaviour, we attempt to improve the quality of the detections using various strategies and show improvement over using vanilla object detection alone. © 2022 IEEE.

Work-in-progress: Identifying unit operations laboratory curriculum needs

A knowledge gap exists between what is taught in the undergraduate chemical engineering curriculum and what is needed in terms of knowledge, skills, and attributes for industry. ABET has worked towards characterizing curriculum needs by specifying seven student outcomes that were developed by their more than 2,200 experts from academia, government and industry. While useful in curriculum assessment, these outcomes are broad and do not provide details about what should be incorporated into specific courses. The National Science Foundation and the American Institute of Chemical Engineers surveyed 507 individuals from industry and academia to identify more specific skills needed for graduates going into industry. In addition to process safety and process dynamic knowledge, results suggested the need for skills in communication, critical thinking, teamwork, leadership, open-ended task analysis, problem solving, and time management. These skills are often taught through the unit operations laboratory. Through the COVID-19 pandemic, faculty were forced to evaluate the learning outcomes of unit operations laboratory courses, allowing them to think more explicitly about how to address those gaps identified through the previous study. Moving forward, faculty have the opportunity to redesign the course to meet the needs of industry. Surveying unit operations laboratory stakeholders will offer a more targeted approach to making the necessary changes in course content. This includes surveying 1. Faculty on their perceptions of the key learning outcomes of the unit operations laboratory, 2. Industry on the knowledge, skills and attributes that should be taught in the laboratory, and 3. Students on what gaps exist in the chemical engineering curriculum that could be filled by the laboratory. The results of these surveys will provide information that will help balance the breadth and depth of content necessary while incorporating updates to address stakeholder needs. This work-in-progress paper will detail the development of these three surveys. © American Society for Engineering Education, 2022.

ENHANCING THE MANAGEMENT OF LONG COVID IN GENERAL PRACTICE: A SCOPING REVIEW

Persistent symptomatic COVID-19 is a multi-system condition that affects approximately 10% of those with acute COVID-19 infection. Affected patients often have complex care needs requiring holistic and multidisciplinary care approaches, the kind routinely provided in general practice. However, there is a lack of evidence of appropriate general practice interventions for the condition. A scoping review was conducted using Arksey and O'Malley's 2005 five-stage framework[1], with later recommendations by Levac et al.[2] to examine the literature and identify knowledge gaps in general practice management of persistent COVID-19. Nineteen papers were selected for review. The studies spanned numerous geographical locations, encompassing several study designs, and a range of populations and sample sizes. The included studies used various definitions for persistent symptomatic COVID-19. The literature was analysed qualitatively, and six major themes were identified. These themes were (i) GP uncertainty, (ii) Listening and empathy, (iii) Assessment and monitoring of symptoms, (iv) Coordinating access to appropriate services, (v) Facilitating provision of continual and integratedmulti-disciplinary care and (vi) Need to facilitate psychological support. Overall, the findings show that general practitioners play a key role in the management of persistent COVID-19, but that more clinical guidance on appropriate interventions is necessary to enhance care. There is a need for scientifically accepted definitions for persistent COVID-19 to ensure that patients can be recognized, assessed and managed appropriately. Patient and public involvement should guide policy makers when developing future care models. Meanwhile, future research should evaluate the implementation and effectiveness of proposed management strategies and interventions.

Emergency Department Utilization Following Statewide Stay at Home Orders During the Novel Coronavirus (COVID-19) Pandemic

Study Objective: With COVID-19 cases and fatalities increasing nationally, health officials implemented policies and restrictions to keep the positivity rate in check. In California, a statewide stay at home order was issued on March 19, 2020 and again on December 7, 2020. The objective of this study was to assess the impact each stay at home order had on ED utilization. Methods: We conducted a multi-center, retrospective study among adult patients (≥18 years) presenting to two emergency departments (urban level 1 trauma center and suburban academic hospital with combined annual census of ∼80,000). We compared ED utilization over a two-week period both prior to and following each statewide stay at home order (March and December, 2020), as well as similar periods in 2019. We calculated the percent change in ED volume and admissions for each. Results: Prior to the first stay at home order, there were only 70 confirmed COVID-19 cases in the county, compared to 94,169 cases prior to the subsequent stay at home order. Compared to 2019, ED volume and admissions during the two-week period following the initial stay at home order decreased by 30.7% and 28.4%, respectively. Following the second stay at home order, ED volume was only down 12.4% from 2019, while admissions were up 4.4%. Similarly, compared to the two week period prior to the initial stay at home order, ED volume and admissions decreased by 22.8% and 14.0% in the following 2-week period, respectively. However, ED volume and admissions remained similar following the second stay at home order, with an increase by 0.8% in ED visits and 1.6% increase in admissions between the two weeks before and after the order began. Conclusion: This study of ED utilization trends during the COVID-19 pandemic demonstrated that ED volume and admissions decreased dramatically during the initial stay at home order. However, despite the large differences in the number county-wide positive cases, ED utilization and admissions were largely unaffected by the second stay at home order.

Investing in a better future: higher education and post-COVID Canada

Post-secondary education (PSE) is a vital part of civil society and any modern economy. When broadly accessible, it can enable socioeconomic mobility, improve health outcomes, advance social cohesion, and support a highly skilled workforce. It yields public benefits not only in improved well-being and economic prosperity, but also in reduced costs in health care and social services. Canada also relies heavily on the PSE sector for research. During the COVID-19 pandemic, PSE has supported research related to the pandemic response and other critical areas, including providing expert advice to support public health and government decision-making, while maintaining educational programs and continuing to contribute to local and regional economies. But the pandemic effort has stretched already strained PSE resources and people even further: for decades, declining public investment has driven increases in tuition and decreases in faculty complement, undermining Canada's research capacity and increasing student debt as well as destabilizing the sector through a growing reliance on volatile international education markets. Given the challenges before us, including climate change, reconciliation, and the pandemic, it is imperative that we better draw on the full range of experience, knowledge, and creativity in Canada and beyond through an inclusive, stable, and globally engaged PSE. Supporting PSE's recovery will be key to Canada's ongoing pandemic response and recovery. The recommendations in this report are guided by a single goal-to make the post-secondary sector a more effective partner and support in building a more equitable, sustainable, and evidence-driven future for Canada, through and beyond the COVID-19 pandemic.

Responding to COVID-19 bereavement in southern ireland: An approach for the whole population

Background: The IHF Bereavement Support Line (BSL) was set up in April 2020 with the HSE to meet anticipated needs of people bereaved by Covid-19 losses. Social support is a strong determinant of psychosocial adjustment following bereavement.1 In a recent study of bereaved people in Australia and Ireland, the majority of respondents (94-80%) reported that family and friends were their main source of support.2 Only between 5% and 19% sought support from professional services. A public health approach to loss recognises that that the normal resources of family, friends, and community are adequate to support the majority of bereaved people in normal circumstances.3 Bereavement guidance recommends a tiered approach to service development.4,5 Level 1/ Universal corresponds to an approach that educates and strengthens public knowledge of bereavement and signposts those in need of intervention to assessment and treatment services. The BSL was designed as an innovative, level 1 service based on Psychological First Aid principles and knowledge of bereavement theory and reactions.6 Its aim was to provide a supportive compassionate listening service, education advice, and signposting to practical, community, and mental health resources. Aim: The aim was to evaluate the pilot phase of the BSL. Method: A reflective learning mixed-methods approach was used. Service data were analysed. All stake holders were interviewed: IHF board members, managers, and staff. Volunteers were surveyed and interviewed to provide insight into their experience of training, supervision, knowledge of models used, skills, personal attitudes, and the cognitive and emotional impact of being a volunteer on the line. A thematic analysis was applied to the qualitative data. Results: Preliminary evidence to date is that the BSL meets a national need, and PFA provides a useful framework for training and delivery. Volunteers find the work challenging but are using supervision and are continuing to volunteer. Full results will be available by November 2020.

70 The Effect of Rapid COVID-19 Testing on Emergency Department Throughput

Study Objectives: Emergency departments (ED) across the world continue to grapple with the COVID-19 pandemic. One growing concern is the ability to rapidly diagnose those infected with the SARS-CoV-2 virus. While rapid assays have proven beneficial in such contexts as strep pharyngitis and influenza, it is unclear whether the recent rapid COVID-19 assays will prove beneficial to ED flow. The purpose of this study is to assess the effect of a rapid COVID-19 assay on patient flow through two academic emergency department sites. Methods: This was a retrospective, multi-facility study conducted between March 10, 2020 and May 9, 2020 at two university hospital EDs that are part of one health system. A rapid COVID-19 assay became available in our health system on April 10, 2020. Included were ED patients of all ages undergoing COVID-19 testing who were considered persons under investigation (PUI). PUIs tested between March 10, 2020 and April 9, 2020 via PCR testing served as the control group. Those tested between April 10, 2020 and May 9, 2020 via the rapid assay comprised the intervention group. Differences in length of stay (LOS) were analyzed between the two groups using T tests and multivariate regression. Results: A total of 9,929 ED patient encounters occurred during the study period, and 3,137 PUIs underwent COVID-19 testing. Average age was 50 years. Fifty-six percent were male. 1,339 PUIs (42.7%) were tested with the PCR test during the control period. 1,798 PUIs (57.3%) were tested with the rapid assay during the intervention period. In the control group, 788 PUIs were discharged and 493 PUIs were admitted. In the intervention group, 512 PUIs were discharged and 1,129 PUIs were admitted. Mean length of stay (LOS) was 341 minutes and 489 minutes for PUI seen in the control period and in the intervention period, respectively (p<.001). When parsed by disposition, differences in mean LOS remained significant for those who were discharged (p<.001), but not for those who were admitted (p=0.35). After controlling for severity index, disposition, and demographic factors, testing with the rapid assay during the intervention period remained associated with an increased length of stay of approximately 95 minutes (95% CI 72-118, p<.001). Conclusion: The use of a rapid COVID-19 assay did not improve patient throughput in our ED and was associated with a longer LOS, especially among those discharged from the ED. Additional testing is needed to determine the utility of the rapid COVID-19 test among an ED population.