Showing papers by "Ryerson University published in 2021"

Behavioral and psychological treatments for chronic insomnia disorder in adults: an American Academy of Sleep Medicine clinical practice guideline.

Abstract: Introduction:This guideline establishes clinical practice recommendations for the use of behavioral and psychological treatments for chronic insomnia disorder in adults.Methods:The American Academy...

Deep learning for object detection and scene perception in self-driving cars: Survey, challenges, and open issues

Abstract: This article presents a comprehensive survey of deep learning applications for object detection and scene perception in autonomous vehicles. Unlike existing review papers, we examine the theory underlying self-driving vehicles from deep learning perspective and current implementations, followed by their critical evaluations. Deep learning is one potential solution for object detection and scene perception problems, which can enable algorithm-driven and data-driven cars. In this article, we aim to bridge the gap between deep learning and self-driving cars through a comprehensive survey. We begin with an introduction to self-driving cars, deep learning, and computer vision followed by an overview of artificial general intelligence. Then, we classify existing powerful deep learning libraries and their role and significance in the growth of deep learning. Finally, we discuss several techniques that address the image perception issues in real-time driving, and critically evaluate recent implementations and tests conducted on self-driving cars. The findings and practices at various stages are summarized to correlate prevalent and futuristic techniques, and the applicability, scalability and feasibility of deep learning to self-driving cars for achieving safe driving without human intervention. Based on the current survey, several recommendations for further research are discussed at the end of this article.

Deep Reinforcement Learning-Based Dynamic Resource Management for Mobile Edge Computing in Industrial Internet of Things

Abstract: Nowadays, driven by the rapid development of smart mobile equipments and 5G network technologies, the application scenarios of Internet of Things (IoT) technology are becoming increasingly widespread. The integration of IoT and industrial manufacturing systems forms the industrial IoT (IIoT). Because of the limitation of resources, such as the computation unit and battery capacity in the IIoT equipments (IIEs), computation-intensive tasks need to be executed in the mobile edge computing (MEC) server. However, the dynamics and continuity of task generation lead to a severe challenge to the management of limited resources in IIoT. In this article, we investigate the dynamic resource management problem of joint power control and computing resource allocation for MEC in IIoT. In order to minimize the long-term average delay of the tasks, the original problem is transformed into a Markov decision process (MDP). Considering the dynamics and continuity of task generation, we propose a deep reinforcement learning-based dynamic resource management (DDRM) algorithm to solve the formulated MDP problem. Our DDRM algorithm exploits the deep deterministic policy gradient and can deal with the high-dimensional continuity of the action and state spaces. Extensive simulation results demonstrate that the DDRM can reduce the long-term average delay of the tasks effectively.

Deep Learning for LiDAR Point Clouds in Autonomous Driving: A Review

Abstract: Recently, the advancement of deep learning (DL) in discriminative feature learning from 3-D LiDAR data has led to rapid development in the field of autonomous driving. However, automated processing uneven, unstructured, noisy, and massive 3-D point clouds are a challenging and tedious task. In this article, we provide a systematic review of existing compelling DL architectures applied in LiDAR point clouds, detailing for specific tasks in autonomous driving, such as segmentation, detection, and classification. Although several published research articles focus on specific topics in computer vision for autonomous vehicles, to date, no general survey on DL applied in LiDAR point clouds for autonomous vehicles exists. Thus, the goal of this article is to narrow the gap in this topic. More than 140 key contributions in the recent five years are summarized in this survey, including the milestone 3-D deep architectures, the remarkable DL applications in 3-D semantic segmentation, object detection, and classification; specific data sets, evaluation metrics, and the state-of-the-art performance. Finally, we conclude the remaining challenges and future researches.

Behavioral and psychological treatments for chronic insomnia disorder in adults: an American Academy of Sleep Medicine systematic review, meta-analysis, and GRADE assessment

Abstract: Introduction:The purpose of this systematic review is to provide supporting evidence for a clinical practice guideline on the use of behavioral and psychological treatments for chronic insomnia dis...

Bipolar DC Power Conversion: State-of-the-Art and Emerging Technologies

Abstract: This article provides a detailed analysis of the power electronics solutions enabling bipolar dc grids. The bipolar dc grid concept has proven to be more efficient, flexible, and higher in quality than the conventional unipolar one. However, despite its many features, these systems still have to overcome their issues with asymmetrical loading to avoid voltage imbalances, besides meeting regulatory and safety requirements that are still under development. Advances in power electronics and the large-scale deployment of dc consumer appliances have put this growing architecture in the spotlight, as it has drawn the attention of different research groups recently. The following provides an insightful discussion regarding the topologies that enable these architectures and their regulatory requirements, besides their features and level of development. In addition, some future trends and challenges in the further development of this technology are discussed to motivate future contributions that address open problems and explore new possibilities.

Spatio-temporal analysis of on-demand transit: A case study of Belleville, Canada

Abstract: The rapid increase in the cyber-physical nature of transportation, availability of GPS data, mobile applications, and effective communication technologies have led to the emergence of On-Demand Transit (ODT) systems. In September 2018, the City of Belleville in Canada started an on-demand public transit pilot project, where the late-night fixed-route (RT 11) was substituted with the ODT providing a real-time ride-hailing service. We present an in-depth analysis of the spatio-temporal demand and supply, level of service, and origin and destination patterns of Belleville ODT users, based on the data collected from September 2018 till May 2019. The independent and combined effects of the demographic characteristics (population density, working-age, and median income) on the ODT trip production and attraction levels were studied using GIS and the K-means machine learning clustering algorithm. The results indicate that ODT trips demand is highest for 11:00 pm–11:45 pm during the weekdays and 8:00 pm–8:30 pm during the weekends. We expect this to be the result of users returning home from work or shopping. Results showed that 39% of the trips were found to have a waiting time of smaller than 15 min, while 28% of trips had a waiting time of 15–30 min. The dissemination areas with higher population density, lower median income, or higher working-age percentages tend to have higher ODT trip attraction levels, except for the dissemination areas that have highly attractive places like commercial areas. For the sustainable deployment of ODT services, we recommend (a) proactively relocating the empty ODT vehicles near the neighbourhoods with high level of activity, (b) dynamically updating the fleet size and location based on the anticipated changes in the spatio-temporal demand, and (c) using medium occupancy vehicles, like vans or minibuses to ensure high level of service.

Green synthesis of carbon dots and their applications

Abstract: Carbon dots (CDs) are nanoparticles with tunable physicochemical and optical properties. Their resistance to photobleaching and relatively low toxicity render them attractive alternatives to fluorescent dyes and heavy metal-based quantum dots in the fields of bioimaging, sensing, catalysis, solar cells, and light-emitting diodes, among others. Moreover, they have garnered considerable attention as they lend themselves to green synthesis methods. Increasingly, one-pot syntheses comprising exclusively of renewable raw materials or renewable refined compounds are gaining favor over traditional approaches that rely on harsh chemicals and energy intensive conditions. The field of green CD synthesis is developing rapidly; however, challenges persist in ensuring the consistency of their properties (e.g., fluorescence quantum yield) relative to conventional preparation methods. This has mostly limited their use to sensing and bioimaging, leaving opportunities for development in optoelectronic applications. Herein, we discuss the most common green CD synthesis and purification methods reported in the literature and the renewable precursors used. The physical, chemical, and optical properties of the resulting green-synthesized CDs are critically reviewed, followed by a detailed description of their applications in sensing, bioimaging, biomedicine, inks, and catalysis. We conclude with an outlook on the future of green CD synthesis. Future research efforts should address the broad knowledge gap between CDs synthesized from renewable versus non-renewable precursors, focusing on discrepancies in their physical, chemical, and optical properties. The development of cost effective, safe, and sustainable green CDs with tunable properties will broaden their implementation in largely untapped applications, which include drug delivery, photovoltaics, catalysis, and more.

Multi-Scale Point-Wise Convolutional Neural Networks for 3D Object Segmentation From LiDAR Point Clouds in Large-Scale Environments

Abstract: Although significant improvement has been achieved in fully autonomous driving and semantic high-definition map (HD) domains, most of the existing 3D point cloud segmentation methods cannot provide high representativeness and remarkable robustness. The principally increasing challenges remain in completely and efficiently extracting high-level 3D point cloud features, specifically in large-scale road environments. This paper provides an end-to-end feature extraction framework for 3D point cloud segmentation by using dynamic point-wise convolutional operations in multiple scales. Compared to existing point cloud segmentation methods that are commonly based on traditional convolutional neural networks (CNNs), our proposed method is less sensitive to data distribution and computational powers. This framework mainly includes four modules. Module I is first designed to construct a revised 3D point-wise convolutional operation. Then, a U-shaped downsampling-upsampling architecture is proposed to leverage both global and local features in multiple scales in Module II. Next, in Module III, high-level local edge features in 3D point neighborhoods are further extracted by using an adaptive graph convolutional neural network based on the K-Nearest Neighbor (KNN) algorithm. Finally, in Module IV, a conditional random field (CRF) algorithm is developed for postprocessing and segmentation result refinement. The proposed method was evaluated on three large-scale LiDAR point cloud datasets in both urban and indoor environments. The experimental results acquired by using different point cloud scenarios indicate our method can achieve state-of-the-art semantic segmentation performance in feature representativeness, segmentation accuracy, and technical robustness.

Assessing the role of quarantine and isolation as control strategies for COVID-19 outbreak: A case study

Abstract: Life style of people almost in every country has been changed with arrival of corona virus. Under the drastic influence of the virus, mathematicians, statisticians, epidemiologists, microbiologists, environmentalists, health providers, and government officials started searching for strategies including mathematical modeling, lock-down, face masks, isolation, quarantine, and social distancing. With quarantine and isolation being the most effective tools, we have formulated a new nonlinear deterministic model based upon ordinary differential equations containing six compartments (susceptible S ( t ) , exposed E ( t ) , quarantined Q ( t ) , infected I ( t ) , isolated J ( t ) and recovered R ( t ) ). The model is found to have positively invariant region whereas equilibrium points of the model are investigated for their local stability with respect to the basic reproductive number R 0 . The computed value of R 0 = 1.31 proves endemic level of the epidemic. Using nonlinear least-squares method and real prevalence of COVID-19 cases in Pakistan, best parameters are obtained and their sensitivity is analyzed. Various simulations are presented to appreciate quarantined and isolated strategies if applied sensibly.

Sources, characteristics, toxicity, and control of ultrafine particles: An overview

Abstract: Air pollution by particulate matter (PM) is one of the main threats to human health, particularly in large cities where pollution levels are continually exceeded. According to their source of emission, geography, and local meteorology, the pollutant particles vary in size and composition. These particles are conditioned to the aerodynamic diameter and thus classified as coarse (2.5–10 μm), fine (0.1–2.5 μm), and ultrafine (

A Public Health Research Agenda for Managing Infodemics: Methods and Results of the First WHO Infodemiology Conference.

Abstract: Background: An infodemic is an overflow of information of varying quality that surges across digital and physical environments during an acute public health event. It leads to confusion, risk-taking, and behaviors that can harm health and lead to erosion of trust in health authorities and public health responses. Owing to the global scale and high stakes of the health emergency, responding to the infodemic related to the pandemic is particularly urgent. Building on diverse research disciplines and expanding the discipline of infodemiology, more evidence-based interventions are needed to design infodemic management interventions and tools and implement them by health emergency responders. Objective: The World Health Organization organized the first global infodemiology conference, entirely online, during June and July 2020, with a follow-up process from August to October 2020, to review current multidisciplinary evidence, interventions, and practices that can be applied to the COVID-19 infodemic response. This resulted in the creation of a public health research agenda for managing infodemics. Methods: As part of the conference, a structured expert judgment synthesis method was used to formulate a public health research agenda. A total of 110 participants represented diverse scientific disciplines from over 35 countries and global public health implementing partners. The conference used a laddered discussion sprint methodology by rotating participant teams, and a managed follow-up process was used to assemble a research agenda based on the discussion and structured expert feedback. This resulted in a five-workstream frame of the research agenda for infodemic management and 166 suggested research questions. The participants then ranked the questions for feasibility and expected public health impact. The expert consensus was summarized in a public health research agenda that included a list of priority research questions. Results: The public health research agenda for infodemic management has five workstreams: (1) measuring and continuously monitoring the impact of infodemics during health emergencies; (2) detecting signals and understanding the spread and risk of infodemics; (3) responding and deploying interventions that mitigate and protect against infodemics and their harmful effects; (4) evaluating infodemic interventions and strengthening the resilience of individuals and communities to infodemics; and (5) promoting the development, adaptation, and application of interventions and toolkits for infodemic management. Each workstream identifies research questions and highlights 49 high priority research questions. Conclusions: Public health authorities need to develop, validate, implement, and adapt tools and interventions for managing infodemics in acute public health events in ways that are appropriate for their countries and contexts. Infodemiology provides a scientific foundation to make this possible. This research agenda proposes a structured framework for targeted investment for the scientific community, policy makers, implementing organizations, and other stakeholders to consider.

A comprehensive review on surface modification of UHMWPE fiber and interfacial properties

Abstract: Ultrahigh molecular weight polyethylene (UHMWPE) fiber is considered as an ideal reinforcing component due to its high strength-to-weight ratio, good toughness and high chemical and wear resistance. However, poor interfacial adhesion with polymer matrices has hindered the development of UHMWPE-based composites with high performance. This review is intended to understand how physical and chemical changes of fiber surface promote the adhesion strengthening mechanisms at the interface and to guide future developments using the presented modifications’ techniques. The review summarized the present state of the art research on surface modification of UHMWPE fiber and UHMWPE/polymer interfacial properties. Various surface modifications of UHMWPE fiber categorized as ‘wet’ chemical and ‘dry’ techniques have been detailed with appropriate examples. Also, the relationship between fiber/matrix adhesion and mechanical properties of the composites has been reviewed. Lastly, an overview of the potential and challenges of each modification method has been discussed.

Strategies for enhancing the room-temperature stretch formability of magnesium alloy sheets: a review

Abstract: Magnesium (Mg) alloys, as one of the lightest structural metallic materials, have attracted considerable attention in the automotive, aerospace, and microelectronic industries. However, wrought Mg alloys are easy to form a strong basal texture with the basal planes of hexagonal close-packed unit cells being parallel to the processing direction during hot processing. This extremely deteriorates the stretch formability of Mg alloy sheets at room temperature (RT) and limits their widespread industrial applications. To overcome this drawback, many studies have been devoted to controlling microstructures including grain sizes, texture characteristics and precipitates to achieve high-performance Mg alloy sheets via alloying and new processing techniques. In this review, we briefly summarize recent advances in enhancing the stretch formability of Mg alloy sheets at RT from two major aspects: (1) by the design of new alloy systems and (2) by the exploitation of advanced processing techniques. Both strategies hold great promise for developing high-performance and low-cost Mg alloy sheets with a superior combination of strength, ductility and stretch formability. Additionally, future research directions for the development of such high-performance Mg alloy sheets are suggested. We hope that this review can provide some insightful information for researchers who are committed to fabricating high-performance Mg alloys for lightweight structural applications in the transportation industry, so as to improve fuel efficiency and reduce climate-changing and health-compromising emissions.

Adaptive Computing Scheduling for Edge-Assisted Autonomous Driving

Abstract: This paper investigates computing resource scheduling for real-time applications in autonomous driving, such as localization and obstacle avoidance. In our considered scenario, autonomous vehicles periodically sense the environment, offload sensor data to an edge server for processing, and receive computing results from the server. Due to mobility and computing latency, a vehicle travels some distance in the duration between the instant of offloading its sensor data and the instant of receiving the computing result. Our objective is finding a scheduling scheme for the edge sever to minimize the above traveled distance of vehicles. The approach is to determine the processing order according to individual vehicle mobility and computing capability of the edge server. We formulate a restless multi-arm bandit (RMAB) problem, design a Whittle index based stochastic scheduling scheme, and determine the index using a deep reinforcement learning (DRL) method. The proposed scheduling scheme avoids the time-consuming policy exploration common in DRL scheduling approaches and makes effectual decisions with low complexity. Extensive simulation results demonstrate that the proposed indexed-based scheme can deliver computing results to the vehicles promptly while adapting to time-variant vehicle mobility.

The role of oil futures intraday information on predicting US stock market volatility

Abstract: This study investigates the role of oil futures price information on forecasting the US stock market volatility using the HAR framework. In-sample results indicate that oil futures intraday information is helpful to increase the predictability. Moreover, compared to the benchmark model, the proposed models improve their predictive ability with the help of oil futures realized volatility. In particular, the multivariate HAR model outperforms the univariate model. Accordingly, considering the contemporaneous connection is useful to predict the US stock market volatility. Furthermore, these findings are consistent across a variety of robust checks.

Numerical and experimental investigation of high concentration aqueous alumina nanofluids in a two and three channel heat exchanger

Abstract: Nanofluids are often able to provide a method to increase the possible heat transfer of a system, with relatively few detrimental factors created by its inclusion. The use of nanofluids and their optimal concentrations has become an area of great interest as of late, with different nanofluid concentrations being key to a systems success or hindrance. The aim of this work is to examine the impact of nanofluid concentration on the possible heat transfer for different heat sink types containing porous media. In this work, both a numerical and experimental investigation into the impact of an aluminum oxide nanofluid on a fluid flow-based system comprised of porous open-cell aluminum foam to examine their impact on the thermal performance of the system was conducted. The porous media implemented in the experimental work was made of a 6061-T6 aluminum with a porosity of 0.91 and a permeability of 2.3869 × 10−7 m2. The experimental work was carried out for a large range of applied heat flux values varying from 3.8328 W cm-2 to 10.3737 W cm−2. The nanofluid examined was also subjected to varying flow rates. The nanofluid concentration was varied from 1% vol and 2% vol Al2O3–water mixture. The nanofluid was suspended in distilled water and mixed using a magnetic mixer. The performance of the work was examined through the Nusselt number to determine the possible thermal enhancement presented by the nanofluid. The use of high concentration (1% vol) nanofluid paired with the use of two different channel designs, both containing porous media resulted in an average thermal enhancement 15% and a maximum enhancement of 24.5% when compared to that of the 2% vol alumina nanofluid.

Machine Learning Assisted Information Management Scheme in Service Concentrated IoT

Abstract: Internet of Things (IoT) has gained significant importance due to its flexibility in integrating communication technologies and smart devices for the ease of service provisioning. IoT services rely on a heterogeneous cloud network for serving user demands ubiquitously. The service data management is a complex task in this heterogeneous environment due to random access and service compositions. In this article, a machine learning aided information management scheme is proposed for handling data to ensure uninterrupted user request service. The neural learning process gains control over service attributes and data response to abruptly assign resources to the incoming requests in the data plane. The learning process operates in the data plane, where requests and responses for service are instantaneous. This facilitates the smoothing of the learning process to decide upon the possible resources and more precise service delivery without duplication. The proposed data management scheme ensures less replication and minimum service response time irrespective of the request and device density.

Medical experimentation and the roots of COVID-19 vaccine hesitancy among Indigenous Peoples in Canada.

Abstract: As the second wave of the pandemic sees case numbers rise to dangerous levels across the country, it has become clear that Indigenous people are particularly vulnerable to coronavirus disease 2019 (COVID-19). The figures released by the Manitoba First Nations COVID-19 Pandemic Response Coordination

STABLE-2007 Demonstrates Predictive and Incremental Validity in Assessing Risk-Relevant Propensities for Sexual Offending: A Meta-Analysis.

Abstract: STABLE-2007 is a measure of risk-relevant propensities for adult males convicted of a sexual offense. This meta-analysis evaluated the ability of STABLE-2007 and its items to discriminate between recidivists and nonrecidivists, and the extent to which STABLE-2007 improves prediction over and above Static-99R. Based on 21 studies (12 unique samples, N = 6,955), we found that STABLE-2007 was significantly and incrementally related to sexual recidivism, violent (nonsexual) recidivism, violent (including sexual) recidivism, and any crime. Scores on STABLE-2007 items and the three STABLE-2000 attitude items also discriminated between individuals who sexually reoffended and those who did not sexually reoffend. These findings support the use of STABLE-2007 in applied risk assessment practice and the interpretation of STABLE-2007 items as indicators of treatment and supervision targets.

ECG Heartbeat Classification Using Multimodal Fusion

Abstract: Electrocardiogram (ECG) is an authoritative source to diagnose and counter critical cardiovascular syndromes such as arrhythmia and myocardial infarction (MI). Current machine learning techniques either depend on manually extracted features or large and complex deep learning networks which merely utilize the 1D ECG signal directly. Since intelligent multimodal fusion can perform at the state-of-the-art level with an efficient deep network, therefore, in this paper, we propose two computationally efficient multimodal fusion frameworks for ECG heart beat classification called Multimodal Image Fusion (MIF) and Multimodal Feature Fusion (MFF). At the input of these frameworks, we convert the raw ECG data into three different images using Gramian Angular Field (GAF), Recurrence Plot (RP) and Markov Transition Field (MTF). In MIF, we first perform image fusion by combining three imaging modalities to create a single image modality which serves as input to the Convolutional Neural Network (CNN). In MFF, we extracted features from penultimate layer of CNNs and fused them to get unique and interdependent information necessary for better performance of classifier. These informational features are finally used to train a Support Vector Machine (SVM) classifier for ECG heart-beat classification. We demonstrate the superiority of the proposed fusion models by performing experiments on PhysioNet’s MIT-BIH dataset for five distinct conditions of arrhythmias which are consistent with the AAMI EC57 protocols and on PTB diagnostics dataset for Myocardial Infarction (MI) classification. We achieved classification accuracy of 99.7% and 99.2% on arrhythmia and MI classification, respectively. Source code at https://github.com/zaamad/ECG-Heartbeat-Classification-Using-Multimodal-Fusion