ABSTRACT
Stringent lockdowns have been one of the defining features of the COVID-19 pandemic. Lockdowns have brought about drastic changes in living styles, including increased residential occupancy and telework practices predicted to last long. The variation in occupancy pattern and energy use needs to be assessed at the household level. Consequently, the new occupancy times will impact the performance of energy efficiency measures. To address these gaps, this work uses a real case study, a two-story residential building in the Okanagan Valley (British Columbia, Canada). Further, steady-state building energy simulations are performed on the HOT2000 tool to evaluate the resiliency of energy efficiency measures under a full lockdown. Three-year monitored energy data is analyzed to study the implications of COVID-19 lockdowns on HVAC and non-HVAC loads at a monthly temporal scale. The results show a marked change in energy use patterns and a higher increase in May 2020 compared to the previous two years. Calibrated energy models built on HOT2000 are then used to study the impacts of pre-COVID-19 (old normal occupancy) and post-COVID-19 (new normal occupancy) on energy upgrades performance. The simulations show that under higher occupancy times, the annual electricity use increased by 16.4%, while natural gas use decreased by 7.6%. The results indicate that overall residential buildings following pre-COVID-19 occupancy schedules had higher energy-saving potential than those with new normal occupancy. In addition, the variation in occupancy and stakeholder preferences directly impact the ranking of energy efficiency measures. Furthermore, this study identifies energy efficiency measures that provide flexibility for the decision-makers by identifying low-cost options feasible under a range of occupancy schedules.
ABSTRACT
COVID-19 is deemed as the most critical world health calamity of the 21st century, leading to dramatic life loss. There is a pressing need to understand the multi-stage dynamics, including transmission routes of the virus and environmental conditions due to the possibility of multiple waves of COVID-19 in the future. In this paper, a systematic examination of the literature is conducted associating the virus-laden-aerosol and transmission of these microparticles into the multimedia environment, including built environments. Particularly, this paper provides a critical review of state-of-the-art modelling tools apt for COVID-19 spread and transmission pathways. GIS-based, risk-based, and artificial intelligence-based tools are discussed for their application in the surveillance and forecasting of COVID-19. Primary environmental factors that act as simulators for the spread of the virus include meteorological variation, low air quality, pollen abundance, and spatial-temporal variation. However, the influence of these environmental factors on COVID-19 spread is still equivocal because of other non-pharmaceutical factors. The limitations of different modelling methods suggest the need for a multidisciplinary approach, including the 'One-Health' concept. Extended One-Health-based decision tools would assist policymakers in making informed decisions such as social gatherings, indoor environment improvement, and COVID-19 risk mitigation by adapting the control measurements.
ABSTRACT
Background: Lung CT scan has a pivotal role in diagnosis and monitoring of COVID-19 patients, and with growing number of affected individuals, the need for artificial intelligence (AI)-based systems for interpretation of CT images is emerging. In current investigation we introduce a new deep learning-based automatic segmentation model for localization of COVID-19 pulmonary lesions. Methods: A total of 2469 CT scan slices, containing 1402 manually segmented abnormal and 1067 normal slices form 55 COVID-19 patients and 41 healthy individuals, were used to train a deep convolutional neural network (CNN) model based on Detectron2, an open-source modular object detection library. A dataset, including 1224 CT slices of 18 COVID-19 patients and 9 healthy individuals, was used to test the model. Results: The accuracy, sensitivity, and specificity of the trained model in marking a single image slice with COVID-19 lesion were 0.954, 0.928, and 0.961, respectively. Considering a threshold of 0.4% for percentage of lung involvement, the model was capable of diagnosing the patients with COVID-19 pneumonia, with a sensitivity of 0.982% and a specificity of 88.5%. Furthermore, the mean Intersection over Union (IoU) index for the test dataset was 0.865. Conclusion: The deep learning-based automatic segmentation method provides an acceptable accuracy in delineation and localization of COVID-19 lesions, assisting the clinicians and researchers for quantification of abnormal findings in chest CT scans. Moreover, instance segmentation is capable of monitoring longitudinal changes of the lesions, which could be beneficial to patients' follow-up.