•  
  •  
 

Abstract

COVID-19, caused by the SARS-CoV-2 virus, was declared a global pandemic by the World Health Organization (WHO) and rapidly spread worldwide from late 2019. While Reverse Transcription Polymerase Chain Reaction (RT-PCR) is the primary diagnostic tool, its sensitivity ranges from only 60% to 70%, leading to false negatives. Computed Tomography (CT) imaging has emerged as a valuable alternative for accurate diagnosis; however, the quality of CT images is often degraded by motion-induced blur and additive noise, particularly in children, individuals with mental health conditions, or those with phobias of CT scans. This study aims to enhance COVID-19 CT image quality using two restoration techniques—one in the spatial domain and one in the frequency domain. The Wiener filter, known for minimizing mean square error and simultaneously addressing noise and blur, is applied and compared with the inverse filtering method. Experimental results using real COVID-19 CT images from Baqubah Teaching Hospital demonstrate that the Wiener filter significantly improves image quality, as confirmed by higher signal-to-noise ratio (SNR) values. It effectively restores images distorted by both motion blur and variable noise, closely approximating the original image content. These enhancements improve the visualization of critical lung features, such as ground-glass opacities, thereby increasing diagnostic accuracy. Overall, the Wiener filter proves to be a reliable and effective method for restoring CT images and aiding in the accurate diagnosis of COVID-19.

Download

Share

COinS