Hybrid Particle Swarm Optimized and Fuzzy C Means Clustering based segmentation technique for investigation of COVID-19 infected chest CT

ABSTRACT

COVID-19 is the world’s most serious threat, affecting billions of people worldwide. Medical imaging, such as CT, has a lot of potential as an alternative to RT-PCR approach for significant judgement and disease control. As a result, automatic image segmentation is in high demand as a therapeutic decision aid. According to studies, medical images may be very useful for early screening since certain aspects of the image imply the existence of virus of COVID-19 and hence may be used as an efficient scanning tool. The proposed work presents a hybrid approach for efficient screening of COVID-19 using chest CT images implementing Hybrid Particle Swarm Optimised-Fuzzy C Means Clustering. The proposed method is tested on 15 chest CT images of COVID-19-infected patients and the results have been validated quantitatively by metrices such as entropy, contrast and standard deviation, which clearly outperforms the conventional Fuzzy C Means Clustering.

KEYWORDS:

• COVID-19
• Particle Swarm Optimization
• Fuzzy C Means Clustering
• image segmentation

Disclosure statement

No potential conflict of interest was reported by the author(s).