ABSTRACT
It has been two years since the global outbreak of the highly contagious and deadly corona virus disease (COVID-19) caused by SARS-CoV-2 first emerged in China. Since then, various diagnostic, prognostic and treatment strategies undertaken to address the pandemic have been dynamically evolving. Predictive and prognostic role of various biomarkers in COVID-19 has been a subject of intense exploration. We aimed to determine the association of Carcinoembryonic antigen (CEA) and various surrogate inflammatory biomarkers with the severity of COVID-19 disease. This retrospective cohort study was carried out on 98 patients admitted in Jaypee Hospital, Noida with COVID-19 disease. Information regarding demographics, laboratory parameters and clinical history was collected from Hospital Information System. Serum levels of CEA and other biomarkers such as Neutrophil-lymphocyte ratio (NLR), C-reactive protein (CRP), Interleukin-6 (IL-6), Ferritin, and Procalcitonin (PCT) were assessed. Correlation analyses were performed between the parameters and acute respiratory distress syndrome (ARDS) stages. Logistic regression and ROC curve analysis were performed to assess the various parameters for distinguishing COVID-19 patients requiring ICU admission. Mean hospital stay, NLR, CEA, IL-6, CRP, Ferritin (P <0.0001) and PCT (P = 0.01) were significantly higher in ICU patients when compared to general ward patients. NLR, median serum CEA, IL-6, and CRP levels were significantly higher in non-survivor compared to the survivors (P <0.0001, 0.0341 and 0.0092). CEA correlated well with disease severity based upon ARDS classification and was a better marker to differentiate patient according to ARDS stages (ARDS 0 vs 2 P = 0.0006;0 vs 3 P <0.0001;ARDS 1 vs 2 P = 0.0183;1 vs 3 P = 0.0006). The area under the Receiver operating characteristic (ROC) curve for CEA was 0.7467 (95% CI-0.64885-0.84459) which revealed the potential of CEA as a biomarker to distinguish COVID-19 patients requiring ICU admission. CEA can be used to predict the severity of COVID-19 associated ARDS as well as patients requiring ICU admission. Along with routine inflammatory biomarkers (NLR, CRP, IL-6, PCT, and ferritin), CEA should be used for early identification of critical COVID-19 positive patients and for assessing prognosis. © 2022, National Institute of Science Communication and Information Resources. All rights reserved.
ABSTRACT
Light and electricity has turned out to be the key part of every individual life. Illumination technology is getting used in different aspects of energy saving and improvises the aesthetic look of the area under the impact of light source. Conventional operation has to be manually ON or OFF, as per our need which means it requires our touch. But in this situation of COVID-19, it has become essential to use smart lighting system technology, because it may not be safe to touch the light switches every time in some places, public areas like offices and hospitals. This major problem can be resolved by using smart lighting system as it is totally based on sensors. In this, LED lights are used to improve the impact of effectiveness. Compared to normal lighting system, the smart lighting system is highly efficient and easy to control. It also saves electricity units consumed. The input of smart lighting system can be either solar panel or conventional grid supply. Use of solar panel involves utilization of solar renewable source. Therefore, the power from the grid dependence is minimized. As it uses solar panels, it can be used in rural areas where there is scarcity of electricity or not yet reach. As per the requirement, even grid supply can be used. In smart lighting system light can be remotely controlled so unnecessary use of power is reduced to some extent, and this makes the system more useful. This paper focus on the reduction of the overall power usage, as well as includes new features with technology interface. © 2021 IEEE.
ABSTRACT
The use of computing technology to aid modern medicine advancement has revolutionized healthcare. Quick and accurate diagnostic tools help professionals start the necessary treatment as soon as possible, saving millions of lives. Pneumonia, a symptom of Covid-19, is a life-threatening condition that affects the lungs and can be detected by analyzing X-ray scans of the chest. The study highlights Convolutional Neural Networks, develops and trains models such as VGG16, ResNet-50, and InceptionV3, to detect pneumonia with improved testing accuracies of 94%, 93.9 %, and 93.5 %, respectively. The work discusses and compares the implementation and performance of these models.