Cluster signal-to-noise analysis for evaluation of the information content in an image

Abstract

Cluster signal-to-noise analysis can simulate the detection performance of observers and thus replace the observer performance test in the evaluation of image quality. Erroneous decision-making increased with increasing background noise.

Prediction of detectability of the mandibular canal by quantitative image quality evaluation using cone beam CT

“…Additionally, our quantitative evaluation method employing the holes to simulate small mass changes could be applied to evaluate not only MCs, but also bone defects or caries in clinical situations. Weerawanich et al 17 developed a new image quality evaluation method that had a higher correlation with observer performance with several cut-off values; however, its relationship to diagnostic performance is still unclear. Although the number of cut-off values (threshold gray value or ΔG) in this method is limited, we confirmed that diagnostic performance was reflected in the number of extracted holes.…”

“…Thus, our method is simpler than Weerawanich's method. 17 There are many comparative studies about diagnostic performance using ROC analysis with several kinds of objectives, while changing scanning conditions or systems. 15,[18][19][20][21][22] ROC analysis is affected by these factors but it eventually derives the result as "diagnostic accuracy".…”