Abstract The coronavirus disease from 2019 (COVID-19) spread over the world in 2020 and caused several health problems. Additionally, because it frequently affects the lungs, automatic detection is particularly crucial for protecting people from death. Using Computed Tomography (CT) images, the Artificial Vulture-based Anamorphic Depth Convolutional (AVbADC) Model is suggested in this study to segment the COVID-19 lungs affected region and categorize COVID-19 cases. Using CT scans of the lungs, the Modified AVbADC model separates COVID-19 infection from other pneumonia cases and normal pneumonia. The suggested architecture is built utilizing two parallel levels with various kernel sizes to capture the local and global properties of the inputs. It is based on the convolutional neural network. The outcomes of the experiment show that our AVbADC. On a short dataset, these results show a promising segmentation and classification performance; more improvements can be made with more training data. All things considered, the updated AVbADC model may be a useful tool for radiologists to aid in the diagnosis and early identification of COVID-19 cases. Finally, the proposed framework's results are contrasted with those of other methods currently in use in terms of sensitivity, accuracy, specificity, F-measure, and other factors.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|