ABSTRACT
Purpose: This paper aims to implement and extend the You Only Live Once (YOLO) algorithm for detection of objects and activities. The advantage of YOLO is that it only runs a neural network once to detect the objects in an image, which is why it is powerful and fast. Cameras are found at many different crossroads and locations, but video processing of the feed through an object detection algorithm allows determining and tracking what is captured. Video Surveillance has many applications such as Car Tracking and tracking of people related to crime prevention. This paper provides exhaustive comparison between the existing methods and proposed method. Proposed method is found to have highest object detection accuracy. Design/methodology/approach: The goal of this research is to develop a deep learning framework to automate the task of analyzing video footage through object detection in images. This framework processes video feed or image frames from CCTV, webcam or a DroidCam, which allows the camera in a mobile phone to be used as a webcam for a laptop. The object detection algorithm, with its model trained on a large data set of images, is able to load in each image given as an input, process the image and determine the categories of the matching objects that it finds. As a proof of concept, this research demonstrates the algorithm on images of several different objects. This research implements and extends the YOLO algorithm for detection of objects and activities. The advantage of YOLO is that it only runs a neural network once to detect the objects in an image, which is why it is powerful and fast. Cameras are found at many different crossroads and locations, but video processing of the feed through an object detection algorithm allows determining and tracking what is captured. For video surveillance of traffic cameras, this has many applications, such as car tracking and person tracking for crime prevention. In this research, the implemented algorithm with the proposed methodology is compared against several different prior existing methods in literature. The proposed method was found to have the highest object detection accuracy for object detection and activity recognition, better than other existing methods. Findings: The results indicate that the proposed deep learning–based model can be implemented in real-time for object detection and activity recognition. The added features of car crash detection, fall detection and social distancing detection can be used to implement a real-time video surveillance system that can help save lives and protect people. Such a real-time video surveillance system could be installed at street and traffic cameras and in CCTV systems. When this system would detect a car crash or a fatal human or pedestrian fall with injury, it can be programmed to send automatic messages to the nearest local police, emergency and fire stations. When this system would detect a social distancing violation, it can be programmed to inform the local authorities or sound an alarm with a warning message to alert the public to maintain their distance and avoid spreading their aerosol particles that may cause the spread of viruses, including the COVID-19 virus. Originality/value: This paper proposes an improved and augmented version of the YOLOv3 model that has been extended to perform activity recognition, such as car crash detection, human fall detection and social distancing detection. The proposed model is based on a deep learning convolutional neural network model used to detect objects in images. The model is trained using the widely used and publicly available Common Objects in Context data set. The proposed model, being an extension of YOLO, can be implemented for real-time object and activity recognition. The proposed model had higher accuracies for both large-scale and all-scale object detection. This proposed model also exceeded all the other previous methods that were compared in extending and augmenting the object detection to activity recognition. The proposed model resulted in the highest accuracy for car crash detection, fall detection and social distancing detection. © 2023, Emerald Publishing Limited.
ABSTRACT
Introduction: The outbreak of coronavirus disease 2019 (COVID-19) was initially detected in Wuhan, China in December 2019. It spread rapidly, and in March 2020, the World Health Organization (WHO) declared a worldwide pandemic. In May 2020, the 73rd World Health Assembly issued a resolution recognizing the role of extensive immunization as a global public-health goal for preventing and stopping transmission of COVID-19. Vaccine hesitancy is a great threat in fighting the COVID-19 pandemic, as it prevents populations from reaching target thresholds of coverage necessary for herd immunity. It is important to know the determinants of vaccine hesitancy so that we can develop tools to combat it. The goal of our study was to evaluate patient perspectives on vaccination in our outpatient Endocrinology clinics. Methods: We created a 7-question survey study which was offered to all patients waiting to be seen in our three Endocrinology clinic locations. We distributed and collected data from all clinics over a 3-week period (5/31/21-6/18/21). We used descriptive statistics to analyze this data. Results: We collected 446 responses between three clinic locations, one urban and two suburban. 361 patients (81%) reported planning to or already having received the COVID-19 vaccine. 29 patients (7%) reported being unsure and 56 patients (13%) reported they did not plan to get vaccinated. Among 85 patients with vaccine hesitancy, 51% are blacks, 35% are whites, 24% reported concerns related to side effects, 13% felt COVID was not as bad as the media portrays, 14% did not believe in vaccines and 29% wanted more data on side effects and efficacy before receiving it. Hesitancy is higher among blacks which is statistically significant (P = 0.035) and slightly higher at the urban compared to the suburban clinics (20% Vs. 15%;p=0.197). On chi square analysis, this had no significant difference. Discussion: In the United States, COVID-19 vaccine acceptance ranged from 60 to 79% in five surveys among the general population. The rate of vaccination acceptance is higher in our study compared to general population. We suspect this is related to our clinic patients having better vaccine counseling as most patients have multiple risk factors for severe infection. The results did show higher vaccine hesitancy among black patients and those seen in our urban clinics suggesting the need for improvement in health literacy in these populations.Presentation: Sunday, June 12, 2022 12:30 p.m. - 2:30 p.m.
ABSTRACT
Objectives: Depression is a leading cause of disability that disproportionately impacts low- and middle-income countries. With greater likelihood of access to a mobile phone than mental health care in most countries, there may be opportunities to leverage these digital tools toward increasing access to depression care. We present a case example from the ESSENCE project in India using digital technology to build capacity of community health workers (CHWs), the essential workforce responsible for frontline maternal and child health services in the country, to deliver a brief evidence-based psychological treatment for depression in primary care. Methods: We discuss key considerations in the design and evaluation of the digital training program, with emphasis on strategies to overcome poor mobile connectivity in rural and underresourced settings, and ensuring participant engagement. We describe the systematic approach to tailor the program to meet the needs of the target user group of CHWs. This involved careful design of the digital program and adapting the training content to the local culture and context. We also describe the use of remote coaching and other support techniques to promote engagement and completion of the training program. Results: We present findings from a pilot study with 42 CHWs that directly informed refinements to the training program and design of a subsequent randomized controlled trial enrolling 340 CHWs. We illustrate the progression from formative research to a larger trial, and the integration of remote coaching support, text messaging, and automated reminders within the digital program to promote engagement. We describe how these efforts are informing the design of a digital platform for training frontline health workers to deliver brief psychological treatments in other settings, including the United States. Conclusions: This example of reciprocal learning in global mental health is particularly timely given the impacts of the COVID-19 pandemic, and renewed efforts to train and support frontline health workers to meet the anticipated surge in mental health challenges globally. This study could yield a blueprint for using widely available digital technology for training and supporting frontline health workers toward scaling up mental health services in India and globally. DDD, TREAT, R
ABSTRACT
Background: Both induction therapy, like oral cladribine, and B-cell depletion therapy, like rituximab, are highly effective disease modulatory treatments (DMTs) in relapsing multiple sclerosis (MS). The high economic costs of the registered DMTs may limit availability of treatment and strain health budgets worldwide. Oral cladribine is a recently approved DMT in Europe, while rituximab is used off-label, especially in Norway and Sweden. Large observational studies indicate good tolerance and treatment effects in MS and studies from other diseases indicate a good safety profile. However, to our knowledge, no phase three studies have compared rituximab with any established highly effective DMT. Formal safety data is also lacking for rituximab treatment in MS. Objectives: To perform a prospective randomized open-label blinded endpoint multicenter non-inferiority study. The primary objective is to test whether rituximab is non-inferior to oral cladribine for treatment of relapsing MS. Methods: In total 264 MS patients with relapsing MS will be recruited from 11 Norwegian centers and followed for 96 weeks. Inclusion criteria are having a relapsing MS diagnosis, age 18-65 years, at least one clinical relapse or one new T2 lesion on MRI within the last year and willingness to use contraception during the study period. Exclusion criteria are contraindications to either treatment, previous use of either or a similar treatment, or treatment with fingolimod or natalizumab (due to risk of rebound activity) within the last six months. The study participants will be treated with either cladribine or rituximab according to current guidelines. Results: The primary endpoint is difference in number of new or enlarging T2 lesions between the two groups from rebaseline at 12 weeks to the end of the study at 96 weeks. Furthermore, we will study clinical course, blood samples and MRI biomarkers to provide tools for personalized MS treatment. Finally, the health economic consequences of these treatment options will be evaluated. At the time of abstract submission, 55 patients have been included across three study sites. The Covid19 outbreak unfortunately resulted in a temporary halt in inclusion from March to May 2020, but the study has now been reopened. End of study is estimated to fall 2023. Conclusions: This study will guide clinicians and patients in future treatment choices for MS. The results will provide valuable knowledge concerning treatment strategies and can potentially have a huge impact on the costs of future MS treatments.