Percentage of ring artifacts correction (PRAC): a quantitative specific evaluation metric for the effect of ring...

X-ray computed tomography (CT) plays an essential role in medical diagnostics, image-guided radiotherapy, and industrial inspection by providing high-resolution cross-sectional images. However, CT image quality can be compromised by ring artifacts-concentric rings that impair image integrity and complicate accurate diagnosis or inspection. Historically, evaluating ring artifact removal has relied on qualitative visual assessments by experts, which, while effective for obvious artifacts, lack reproducibility when identifying subtle structures. Traditional image quality metrics are not designed to assessing ring artifact correction and typically require a reference image, which is often unavailable in practical applications. To address these challenges, this study introduces Percentage of Ring Artifact Correction (PRAC), a metric designed specifically to evaluate the effectiveness of ring artifact correction in X-ray CT images without needing a reference image. PRAC quantifies artifact reduction by measuring the percentage decrease in standard deviation along the radial direction within a polar coordinate system, thus providing an accurate measure of artifact correction. Validation against established metrics like Root Mean Square Error (RMSE), Signal-to-Noise Ratio (SNR), Structural Similarity Index (SSIM), and Contrast-to-Noise Ratio (CNR) demonstrates PRAC's stability under varying noise conditions, sensitivity to minor variations, and correlation with ring artifact intensity. Our findings confirm PRAC's efficacy as a specialized metric for assessing ring artifact correction in CT images, reliably distinguishing the impact of various artifact removal algorithms, thereby enhancing the accuracy and reliability of CT imaging through a robust evaluation framework.