A Novel Cnn Based Smart Mask Detection System

The Coronavirus made a new normalization of life where communal distancing and use of masks for covering their face perform an essential part in monitoring the effects of spreading of the corona virus, still the majority of population are found not using face shields or masks in public areas that accelerates the spreading of the corona virus. This might lead to the serious issue of rise in scattering of the disease. Therefore, to neglect any kind of circumstances we are in need to explore and alert the public for wearing masks. Persons can't be deployed for this procedure, as the risk of getting affected by corona virus increases. Henceforth, the presented model for mask detection is surrounded along the theories of artificial intelligence (AI), deep learning, object detection technologies and convolutional neural networks (CNN) which are the key subject of this project. The project performs by recognizing the people are wearing their face shields or masks or not in public areas via utilizing image processing and deep learning practices and transmitting data to the governing authorities. These algorithms for abject detection have been optimized for recognition of people with face masks or not. This paper is attempting for development of a model for real-time monitoring which will turn out to be pretty effective and simple. This model magnificently recognizes whether an individual is wearing a mask or not up to 98% of accuracy as achieved till date and observed that it has yielded outstanding outcomes for the detection. Copyright © 2022 Wolters Kluwer Medknow Publications. All rights reserved.

Acnn Basedsmartmask Detectionsystem

The Coronavirus made a new normalization of life where communal distancing and use ofmasks for covering their face perform an essential part in monitoring the effects of spreading of thecorona virus, still the majority of population are found not using face shields or masks in public areasthat accelerates the spreading of the corona virus. This might lead to the serious issue of rise inscattering of the disease. Therefore, to neglect any kind of circumstances we are in need to exploreand alert the public for wearing masks. Persons can't be deployed for this procedure, as the risk ofgetting affected by corona virus increases. Henceforth, the presented model for mask detection is surroundedalongthetheoriesofartificialintelligence(AI), deep learning, object detection technologies andconvolutionalneuralnetworks(CNN)whicharethekeysubjectofthisproject.Theprojectperformsbyrecognizing the people are wearing their face shields or masks or not in public areas via utilizingimage processing and deep learning practices and transmitting data to the governing authorities.These algorithms for abject detection have been optimized for recognition of people with face masksor not. This paper is attempting for development of a model for real-time monitoring which will turnout to be pretty effective and simple. This model magnificently recognizes whether an individual iswearing amask or not up to 98% of accuracy as achieved till date and observed that it has yieldedoutstandingoutcomesforthedetection. Copyright © 2022, Anka Publishers. All rights reserved.

REMDESIVIR THERAPY AND ITS ASSOCIATION WITH SINUS BRADYCARDIA

TOPIC: Chest Infections TYPE: Original Investigations PURPOSE: Various case reports have associated transient sinus bradycardia with Remdesivir (RDV) therapy for SARS-CoV2 infection. Pallotto et al., Gubitosa et al., and Touafchia et al. all reported the association of Remdesivir with increased risk of bradycardia in small samples of patients treated with RDV for COVID-19. To our knowledge, no large studies looked at this side effect of RDV therapy. We aimed to analyze the association between sinus bradycardia and Remdesivir therapy for COVID-19. METHODS: A retrospective case-control analysis was done for 1535 patients with SARS-CoV2 infection who were admitted to four teaching hospitals in an urban area in 2020. The mean age was 66 years (SD of 16.7, range 18-99), with 774 males (50.4%). Patients were divided into cases (treated with RDV) and controls (not treated with RDV). Multivariate logistic regression methods were used to analyze the associations between independent variables and outcomes. Pulse rate variables were recorded as pulse rate at day-0, day-3, day-7, and incidence of bradycardia on three consecutive days during admission. Other variables recorded were age, gender, comorbidities, prior history of cardiac disease/arrhythmias, concomitant medications (including AV nodal blockers, dexamethasone, Albuterol, and Lasix), and ICU admission. Survival analysis was run for 7-day and 30-day mortality, as well as survival to hospital discharge. RESULTS: 1415 patients were included in the final analysis, after the exclusion of 120 patients with previous heart blocks. 600 patients (39.1%) were in the Remdesivir group, and 935 patients (60.9%) were in the control group. Between both groups, a total of 454 patients (29.6%) had transient bradycardia on three consecutive days during hospitalization. A multivariate regression analysis was done after adjusting for all confounding variables (age, gender, history of cardiac diseases, AV-nodal blocking drugs, dexamethasone, furosemide, and albuterol therapy). It was seen that there was no statistically significant association between RDV therapy and persistent transient bradycardia (transient bradycardic events on three consecutive days) (Odds Ratio 0.823, 95% confidence interval (CI) 0.594-1.134, p=0.236). There was no statistically significant association of RDV therapy with patients having any bradycardia event during hospitalization in a sub-analysis (Odds Ratio 0.888, 95% confidence interval (CI) 0.665-1.184, p=0.419). Also, RDV failed to show any statistically significant mortality benefit (OR 1.1, CI 0.75-1.62, p=0.6). CONCLUSIONS: Our findings indicate that although transient sinus bradycardia in patients with COVID-19, can be triggered by severe hypoxia, inflammatory damage to AV-nodal cells, or exaggerated response to medications, there was no statistically significant association of RDV therapy with the risk of developing bradycardia. RDV therapy should not be withheld in patients at risk of developing bradycardia. CLINICAL IMPLICATIONS: Remdesivir therapy did not increase the risk of developing bradycardia in our patient population. RDV therapy should not be withheld in patients at risk of developing bradycardia. Larger RCTs are needed to validate these findings. DISCLOSURES: No relevant relationships by Rahul Bollam, source=Web Response No relevant relationships by Bhagat Kondaveeti, source=Web Response No relevant relationships by Florencio Mamauag, source=Web Response No relevant relationships by Kainat Saleem, source=Web Response No relevant relationships by Manasi Sejpal, source=Web Response No relevant relationships by Megha Sood, source=Web Response No relevant relationships by Morgan Stalder, source=Web Response No relevant relationships by Rosalie Traficante, source=Web Response No relevant relationships by Syed Arsalan Zaidi, source=Web Response

Student performance prediction for adaptive e-learning systems

The growth of e-learning systems has changed current learning behavior and tries to present a new framework for the learners. E-learning platforms have become common and approachable for a vast set of audiences. The COVID-19 pandemic in 2020 has triggered the application of these online learning platforms. The number of e-learning platforms has been increasing rapidly to fulfill the requirement. This chapter tries to estimate the three factors consisting of learner’s personality, learning style and knowledge level in order to recommend the content that is best suited to the learner. An ensemble approach to solving this problem has been used, which utilizes a genetic algorithm and KNN to find the content appropriate for the learner. © The Institution of Engineering and Technology 2021.

Implementation of SimpleRNN and LSTMs based prediction model for coronavirus disease (Covid-19)

Deep learning is a powerful technique which is inspired by the structure as well as processing power of the human brain. This technique uses deep neural network to perform complex tasks such as time series prediction, image classification, and cancer detection. In this research work, we used Covid-19 time series datasets and with the help of deep learning we built the model for prediction of Covid-19 cases. For the model building, we used two deep learning neural networks, Recurrent Neural Networks (RNN) and Long Short Term Memory Networks (LSTMs). We built a prediction model using RNN in the first instance and subsequently the second model was built using LSTMs. Out of these two neural networks, we got promising results from the model based on LSTMs with an overall accuracy of 98% . As the cases of Covid-19 are increasing day-by-day at a very high rate, we proposed these models using neural networks to help in predicting the future trends of Covid-19 confirmed, deaths and recovered cases. © 2021 Institute of Physics Publishing. All rights reserved.

St Andrew's COVID-19 Surgery Safety (StACS) study: Skin cancer.

BACKGROUND: Skin cancer represents the most common malignancy worldwide and it is imperative that we develop strategies to ensure safe and sustained delivery of cancer care which are resilient to the ongoing impact of COVID-19. OBJECTIVE: This study prospectively evaluates the COVID-19 related patient risk and skin cancer management at a single tertiary referral centre, which rapidly implemented national COVID-19 safety guidelines. METHOD: A prospective cohort study was performed in all patients who underwent surgery for elective skin cancer service management, during the UK COVID-19 pandemic peak (April-May 2020). 'Real-time' 30-day hospital database deceased data were collected. Random selection was undertaken for patients who either underwent operative (surgery group) management or remained on the waiting list (control group); these groups were also prospectively followed-up within a controlled cohort study design and telephoned at the end of June 2020 for the control group or 30 days post-operatively. RESULTS: Of the 767 patients who had operations, there were no COVID-19 related deaths. Both the surgery (n = 384) and control (n = 100) groups were matched for age, sex, ethnicity, BMI, presence of comorbidities, smoking and positive COVID-19 contact. There were no differences in post-operative versus any symptom development (1.3%, 5/384 vs. 4%, 4/100, p = 0.093), or proportion of positive tests (8.6%, 33/384 vs. 8%, 8/100; p = 0.849), between the surgery and control groups. CONCLUSION: These data support continued and safe service provision, and no increased risk to skin cancer patients who require surgical management, which is vital for continuation of cancer treatment in the context of a pandemic. LEVEL OF EVIDENCE: II.

St Andrew's COVID-19 surgery safety study: hand trauma.

INTRODUCTION: Adaptation is vital to ensure successful healthcare recovery during the COVID-19 pandemic. Hand trauma represents the most common acute emergency department presentation internationally. This study prospectively evaluates the COVID-19 related patient risk, when undergoing management within one of the largest specialist tertiary referral centres in Europe, which rapidly implemented national COVID-19 safety guidelines. MATERIALS AND METHODS: A prospective cohort study was undertaken in all patients referred to the integrated hand trauma service, during the UK COVID-19 pandemic peak (April-May 2020); all were evaluated for 30-day COVID-19 related death. Random selection was undertaken for patients with hand trauma who either underwent non-operative (control group) or operative (surgery group) management; these groups were prospectively followed-up within a controlled cohort study design and telephoned at 30 days following first intervention (control group) or postoperatively (surgery group). RESULTS: Of 731 referred patients (566 operations), there were no COVID-19 related deaths. Both groups were matched for sex, age, ethnicity, body mass index, comorbidities, smoking, preoperative/first assessment COVID-19 symptoms, pre- and postoperative/first assessment isolation and positive COVID-19 contact (p > 0.050). There were no differences in high service satisfaction (10/10 compared with 10/10; p = 0.067) and treatment outcome (10/10 compared with 10/10; p = 0.961) scores, postoperative/first assessment symptoms (1%, 1/100 compared with 0.8%, 2/250; p = 1.000) or proportion of positive tests (7.1%, 1/14 compared with 2.2%, 2/92; p = 0.349), between the control (n = 100) and surgery (n = 250) groups. CONCLUSION: These data support continued and safe service provision and no increased risk to patients who require surgical management. Such findings are vital for healthcare providers when considering service adaptations to reinstate patient treatment.

St Andrew's COVID-19 surgery safety (StACS) study: Elective plastic surgery, trauma & burns.

INTRODUCTION: This study evaluates COVID-19 related patient risk, when undergoing management within one of the largest specialist centres in Europe, which rapidly implemented national COVID-19 safety guidelines. METHOD: A prospective cohort study was undertaken in all patients who underwent surgical (n = 1429) or non-operative (n = 191) management during the UK COVID-19 pandemic peak (April-May 2020); all were evaluated for 30-day COVID-19 related death. A representative sample of elective/trauma/burns patients (surgery group, n = 729) were selected and also sub-analysed within a controlled cohort study design. Comparison was made to a random selection of non-operatively managed (non-operative group, n = 100) or waiting list (control group, n = 250) patients. These groups were prospectively followed-up and telephoned from the end of June (control group) or at 30 days post-first assessment (non-operative group)/post-operatively (surgery group). RESULTS: Complex general (9.2%, 136/1483) or regional (5.0%, 74/1483) anaesthesia cases represented 14.2% (210/1483) of operations undertaken. There were no 30-day post-operative (0/1429)/first assessment (0/191) COVID-19 related deaths. Neither the three sub-speciality plastic surgery, or non-operative groups, displayed increases in post-operative/first assessment symptoms in comparison to each other, or to control. The proportion of COVID-19 positive tests were: 7.1% (1/14) (non-operative), 5.9% (2/34) (burns) and 3.0% (3/99) (trauma); there were however no significant differences between these groups, the elective (0%, 0/54) and control (0%, 0/24) groups (p = 0.236). CONCLUSION: We demonstrate that even heterogeneous sub-speciality patient groups, who required operative/non-operative management, did not incur an increased COVID-19 risk compared to each other or to control. These highly encouraging results were achieved with described, rapidly implemented service changes that were tailored to protect each patient group and staff.