Detecting Subacute Thyroiditis after COVID-19 infection using Deep Learning Techniques – A Case Study. Issue 1 (June 2021)

• Record Type: Journal Article
• Authors: Abdullah, A Sheik; Suganya, R; Abirami, A M; Bhubesh, K R A
• Is Part Of: Journal of physics. Volume 1917:Issue 1(2021)
• Publication Date: 2021-06
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1917/1/012023
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: The act of Subacute Thyroiditis (ST) is considered to be a challenge in medical informatics which severely affects half of everybody. It plans the thyroid knobs of high-goal thyroid ultrasound. The mechanism formulates the C cells in the Thyroid gland, which makes a hormone that helps control the levels of calcium in the body which gets affected after COVID-19. Hundreds of research teams have been working in recent days to collect data and apply computational techniques to analyze and interpret the experimental results in COVID-19. Still, only a few researched the impact of Thyroid due to COVID-19. In the proposed architecture, we introduced CNN with GapNET-PL and Convolutional Block Attention Module (CBAM) to improve the overall mechanism and also utilize the significant features from the Pooling layer. We collected real-time ultrasound thyroid image dataset consists of 19 images from COVID-19 infected patients for preparing and approval of the proposed model. The significance of this work is to gauge the relationship between's contamination with COVID 19 and the improvement of ST utilizing AI procedures. The trial results show that the assessment of thyroid datasets with the proposed approach has given 89% of accuracy with improvement in P-value <0.001. The training parameters provided valid results with improvisation in statistical performance. Thereby this research can support doctors in the domain of imaging analytics with the aid of AI-systems in reasoning COVID-19 diseases related to Subacute Thyroiditis. Hence this can be strongly recommended for the validation of medical data and its risk factors that contribute towards the disease.