Title: Deep Learning for COVID-19 Lung CT Image Segmentation: State-of-the-Art and Future Directions  

Authors: Mrs. Varalaxmi Adimurthy, Assistant Professor , Department of Computer Science , Government First Grade College, Kadugudi, Bangalore -560067

Abstract: The COVID-19 pandemic has had a devastating impact on the global population, and early diagnosis and treatment are essential to reducing mortality and morbidity. Computed tomography (CT) imaging is a valuable tool for diagnosing COVID-19, and lung CT image segmentation is a promising technique for automatically identifying and quantifying the infected regions. In this paper, we review the state-of-the-art methods for COVID-19 CT lung image segmentation. We discuss the challenges of this task, including the variability of CT images, the presence of artifacts, and the need for high accuracy. We then review a number of deep learning-based methods that have been proposed for COVID-19 CT lung image segmentation. These methods have achieved promising results, and they have the potential to automate the diagnosis and management of COVID-19. We conclude by discussing the future directions of research in COVID-19 CT lung image segmentation. We highlight the need for larger and more diverse datasets, the development of more robust and interpretable methods, and the integration of segmentation with other tasks, such as diagnosis and prognosis.

Keywords: COVID-19, CT Lung Image Segmentation, Deep Learning, Image Segmentation

DOI: 10.5281/zenodo.8199293

International Journal of Applied Pattern Recognition, 2023 Vol.7 No.3, pp.194 - 197

Received: 15 Dec 2022
Accepted: 18 March 2023
Published online: 27 June 2023