• Kirtika, Dr. Rajinder Singh Sodhi


The COVID-19 outbreak has had a momentous influence on worldwide public health, thereby requiring efficacious tactics for timely identification, identification confirmation, and treatment. This scholarly article presents a comprehensive survey of the significance of chest computed tomography (CT) imaging in the detection of COVID-19 and accentuates the benefits of deep learning in scrutinising CT images for the diagnosis of COVID-19. A plethora of deep learning structures, such as U-Net, ResNet, and DenseNet, are comprehensively expounded, alongside their modifications for the identification of COVID-19. The present discourse expounds on the evaluation metrics utilised for the appraisal of the efficacy of deep learning models. This study delves into an analysis of the obstacles and constraints associated with the identification of COVID-19 through deep learning methodologies. Additionally, the significance of collaborative efforts among multiple centres and the sharing of data to enhance the efficacy of the model are underscored. The ethical implications and possible prejudices associated with the implementation of deep learning algorithms for the identification of COVID-19 are deliberated, underscoring the necessity for openness and impartiality. This study delves into potential avenues for future research, including the examination of multimodal approaches and longitudinal analysis. Additionally, ethical considerations and potential biases linked to deep learning models are scrutinised. The article culminates by emphasising the prospective employment of profound learning in transforming the diagnosis of COVID-19 and augmenting the efficacious management of the disease. Keywords: COVID-19, deep learning, CT imaging, convolutional neural networks, U-Net, ResNet, DenseNet, evaluation metrics, challenges, limitations, multi-center collaborations, data sharing, ethics, biases, transparency, future directions.
How to Cite
Kirtika, Dr. Rajinder Singh Sodhi. (1). COVID-19 DETECTION IN CT IMAGES WITH DEEP LEARNING: ADVANCEMENTS, CHALLENGES, AND FUTURE DIRECTIONS. International Journal Of Innovation In Engineering Research & Management UGC APPROVED NO. 48708, EFI 5.89, WORLD SCINTIFIC IF 6.33, 10(3), 01-07. Retrieved from