RESUMEN
Background: As of December 2022, SARS-CoV-2 coronavirus resulted in over 6 million deaths worldwide.[1] It was realized early into the pandemic, that COVID-19 significantly impacts the Cardiovascular system. [2] Patients with pre-existing cardiovascular comorbidities were particularly at higher risk of adverse outcomes during their hospitalizations. [3] Similarly, it can be safe to assume patients with adult congenital heart disease (ACHD) should considered a high-risk population for the development of severe COVID infection with increased rates of significant cardiovascular complications. Objective(s): Based on this reasoning and the paucity of data available on this topic using a large database, we sought to investigate the outcomes of patients with ACHD who were admitted to the hospital with COVID-19. Method(s): The National Inpatient Sample database for 2020 was queried to identify adult hospitalizations with a primary diagnosis of COVID-19 and a secondary diagnosis of ACHD using International Classification of Diseases - 10 Clinical Modification (ICD-10-CM) codes. The primary outcome studied was inpatient mortality, while secondary outcomes included inpatient complications, mean length of stay (LOS), and total hospital charge (THC). Multivariate logistic and linear regression analyses were used to adjust for possible confounders and analyze the variables. Result(s): Out of 1,050,045 COVID-19 hospitalizations registered, 2,425 (0.23%) had ACHD as a secondary diagnosis. Encounters with ACHD who were hospitalized with COVID-19 had significantly higher adjusted odds of inpatient mortality (Adjusted Odds Ratio [aOR]: 1.4, [95% CI: 1.05-1.88], p=0.022), Longer LOS (Mean 2.4 days, [95% CI: 1.35-3.40], p <0.001), and higher Total Hospital Charges (Mean $53,000, [95% CI: 20811-85190], p <0.001). A Forrest plot (Figure 1) demonstrates a graphical representation of the multivariate analysis of the significant in-hospital complications when adjusted for patient demographics, comorbidities, and hospital characteristics. Conclusion(s): Among COVID-19 hospitalizations, those with a history of congenital heart diseases had significantly worse outcomes in terms of in-hospital mortality, sepsis;the need for endotracheal intubation, mechanical ventilation, and vasopressors;developing acute kidney injury and pulmonary embolism, in addition to the longer length of stay, and higher total hospital charges. [Formula presented]Copyright © 2023
RESUMEN
Corona Virus Disease (COVID-19) has hit the world hard and almost every country has faced its consequences may be the population and number of people affected or economically. Crowd management is incredibly tough for big surroundings and continuous watching manually is troublesome to execute. Vaccinated people are also getting affected by the virus so it is advisable to take Public Health & Social Measures (PHSM) such as wearing a proper mask, sanitization and keeping social distancing in crowded places. The proposed paper presents a machine learning based real-time Covid alert and prevention system to ensure Covid appropriate behavior in public places and social gatherings. There are three modules under this system: (i) Real-time Face mask detection, where persons with masks, improper masks or no mask are detected and classified;(ii) Real-time people counting for ensuring a limit on public meetings and social gatherings and (iii) Real-time social distance monitoring. All these modules are integrated and deployed on embedded hardware, NVidia's Jetson Nano. The implementation results are presented and analysis of the detection is done in real-time on the edge-AI platform. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
RESUMEN
The pandemic of the coronavirus COVID-19 is having an effect on public health around the globe. The WHO has advised a number of precautions to control the spread of the virus, one of which is the proper usage of a face mask. However, many people are violating this simple measure in public places, which could increase the risk of virus transmission. This inspires us to develop a face mask detection system using deep learning to identify whether individuals are wearing masks and how well they are wearing them in real-time. The system is proposed with two cutting-edge object detection models, YOLOv3 Tiny and YOLOv4 Tiny. This system is capable of detecting and classifying face masks as well as their wearing condition and provides class-wise counting in real-time. For real-time inference, these models were deployed on NVIDIA Jetson Nano hardware. The detection models obtained mAP of 84.2% and 88.4% and FPS of 23.94 and 23.58 for YOLOv3 Tiny and YOLOv4 Tiny respectively. In terms of mAP, YOLOv4 Tiny performed better than YOLOv3 Tiny while both have nearly the same FPS, thereby giving an efficient mask detection in real-time, making it a deployable system for the public or crowded places. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
RESUMEN
The coronavirus pandemic (COVID-19) has unfolded hastily throughout the entire world. This pandemic disease can spread through droplets and can be airborne. Hence, the use of face masks in public places is crucial to stop its spread. The present study aims to develop a system that can identify masked or non-masked faces;whether it is a normal mask, transparent mask, or a face alike mask. The face mask detection system is developed with the help of Convolutional Neural Networks (CNN). The model compression technique of Knowledge Distillation has been used to make the machine lesser computation and memory intensive so that it is simple to install the model on a few embedded gadgets and cell computing platforms. Using the model compression technique and GPU systems will help boom the calculation velocity of the model and drop the storage space required for calculations. The experimental outcomes show that the developed detector is capable to classify diverse types of masks. Also, it can classify video images in real-time. Using the Knowledge Distillation on the baseline model can improve the testing accuracy from 88.79% to 90.13%. The proposed unique system can be implemented to assist in the prevention of COVID-19 spread and detect various mask types. © 2021 IEEE.
RESUMEN
Real-time face mask detection with the use of Artificial Intelligence is one of the most advanced ways of detecting face masks and their wearing condition in public or private areas. In this work, a system based on Object Detection models is proposed which can detect and classify the type of mask wearing conditions in real-time. The system is implemented with two latest deep convolutional neural networks;YOLOv5s and YOLOv5l. The proposed system can efficiently detect and classify face masks based on their wearing condition as well as count them and store the count into a CSV file format with a timestamp. To perform real-time inference, the deep learning models were deployed on Nvidia Jetson Nano and Jetson Xavier NX which are embedded solutions inspired by Edge AI. The detection algorithms achieved mAP of 86.43 and 92.49 for YOLOv5s and YOLOv5l respectively. Comparing the mAP of both detection models, YOLOv5l achieved higher mAP than YOLOv5s while comparing fps on both hardware, Nvidia Jetson Xavier NX provides more fps than Nvidia Jestion Nano for realtime inference. © 2021 IEEE.
RESUMEN
The vaccination drive for the much dangerous and contagious Coronavirus (COVID-19) has started successfully in India. This paper proposes to predict the vaccination drive of COVID-19 using the time series data for India. The proposed model was used for predicting the number of people to be vaccinated once per day in the country. The proposed model was compared with the direct input-based Long Short Term Memory (LSTM) cell model using various performance parameters and the proposed model was found to perform better. The actual closeness of the model's prediction from the actual data was depicted through line graphs. The proposed model was further used to predict the short-term and long-term future values. Herd immunity is another key ongoing research area when it comes to COVID-19. The Herd Immunity Threshold (HIT) of COVID-19 has not been found yet. However, this paper has proposed the expected number of days for different population thresholds. The proposed model predicts 174 days for obtaining a population threshold of 50% and 319 days for obtaining a population threshold of 90%. © 2021 IEEE.