Comparative study of COVID and pulmonary fibrotic CT lung images using siamese networks with VGG16 / S. V. Aksenov, N. J. Samuel Ragland Francis, N. S. Samuel Ragland Francis
Язык: английский.Страна: Россия.Серия: Искусственный интеллект и машинное обучениеРезюме или реферат: In this research an algorithm is proposed to produce comparative results between Pulmonary Fibrosis of the lungs and COVID computer tomography lung images for the purpose of research to aid in the field of medical science. The Siamese Network which is based on parallel tandem operation to produce comparative results, is altered by changing or altering the implementation function using the VGG16 neural network. The input data set in the method uses a variation of healthy lung CT images along with CT images of cases with pulmonary fibrosis and COVID. The main aim is to produce a comparative study on the textural variation of the CT images under study to further enhance research outputs in the future with accuracy and less time consumption..Примечания о наличии в документе библиографии/указателя: [Библиогр.: с. 101 (18 назв.)].Тематика: электронный ресурс | труды учёных ТПУ | компьютерная томография | COVID-19 | нейронные сети | медицинские изображения | диагностические исследования Ресурсы он-лайн:Щелкните здесь для доступа в онлайнЗаглавие с титульного экрана
[Библиогр.: с. 101 (18 назв.)]
In this research an algorithm is proposed to produce comparative results between Pulmonary Fibrosis of the lungs and COVID computer tomography lung images for the purpose of research to aid in the field of medical science. The Siamese Network which is based on parallel tandem operation to produce comparative results, is altered by changing or altering the implementation function using the VGG16 neural network. The input data set in the method uses a variation of healthy lung CT images along with CT images of cases with pulmonary fibrosis and COVID. The main aim is to produce a comparative study on the textural variation of the CT images under study to further enhance research outputs in the future with accuracy and less time consumption.
Для данного заглавия нет комментариев.