ABSTRACT
The COVID-19 outbreak and subsequent public health interventions have depressed demand and disrupted supply chains for many fishing businesses. This paper provides an analysis of the COVID-19 impacts on the profitability of the EU fishing fleets. Nowcasting techniques were used to estimate the impact of the COVID-19 pandemic on the economic performance for the EU fishing fleet in 2020 and 2021. Our results show that the economic impact of COVID-19 on this sector was smaller than initially expected and overall profits remained positive. This was in part due to low fuel prices that reduced operating costs of fishing, and the early response from governments to support the sector. The results vary by fishing fleet, revealing that small-scale fleets and the fleets in the Mediterranean and Black seas have been more impacted than large-scale fleets and the fleets in the Northeast Atlantic. © G. Carpenter et al., Published by EDP Sciences 2023.
ABSTRACT
Covid-19 pandemic is a major health thread all over the world. Early detection is the only solution to control the spread of disease. Chest X-rays plays a key role in the diagnosis of Covid-19 since the viral test and the antibody test may take time to get the result. These tests sometimes give the result negative for infected persons. Chest X-rays are also cost effective when compared to other diagnosis tests for Covid-19 patients. Medical image analysis requires more efforts as the data increases rapidly. Due to high risk of work in this area, a Computer aided technique can lead to diagnose Covid-19 accurately than the radiologist. Better solution is to use machine learning techniques for risk assessment and treatment planning. This model can classify Covid-19 patients, Pneumonia patients and healthy patients based on their chest X-rays. Statistical measures are used in machine learning to retrieve the hidden information present in the image that may be used for good decision-making. X-ray images are gray scale images with almost the same textural features. In our model the traditional textural feature Gray Level Co-occurrence matrix (GLCM) is used to extract the information of pixel intensities between neighbouring pixels in a small region in the chest X-ray images of the patients. Then these extracted features of the patients are given to different conventional machine learning techniques like K-Nearest neighbor, Naive-Bayes Classifier, Support Vector machine for classification. Comparison of these classifiers are done on the basis of accuracy and found to be less. Then advanced machine learning ensemble methods were tried for classification. The ensemble methods like Random forest and XGBoost are used for classification. The comparative study of the model shows that classifying the X-ray image dataset with the combination of GLCM and ensemble methods gives better result than using GLCM with traditional machine learning methods. Our model has less computation time and it requires less memory (cost effective).