A new image quality measure in CT: Feasibility of a contrast-detail measurement method

Abstract

PurposeTo develop a new method of evaluating image quality in computed tomography (CT) using an objective measure of low contrast-detail (LCD). MethodTo achieve this aim a new LCD-CT (CDCT) phantom needed to be designed and developed. A CT inverse image quality figure (CT-IQFinv) value, based on the planar radiography LCD method, was also devised. Validation of the CDCT phantom design and CT-IQFinv calculations were undertaken using 67 observers and software methods. The CDCT phantom was scanned on three multi-detector CT systems using variable factors of kVp, mAs and slice thickness. ResultsThe results were compared to an a priori knowledge that image quality improves with increased photons reaching the detectors. Observer CT-IQFinv scores for the phantom's peripheral region were consistent with the a priori knowledge and generally consistent in the inner region, with one exception. The software CT-IQFinv scores for the phantom's peripheral region were also consistent with the a priori knowledge, however there were some inconsistencies. Software and observer CT-IQFinv score differed significantly (p < 0.05) however both were consistent with the a priori knowledge. ConclusionsThe work reported is designed as proof of concept of development of LCD measure in CT. CT-IQFinv can be used as a measure of LCD image quality in CT when evaluating CT parameter of mAs, kVp and slice thickness. The results demonstrate potential for use of CT IQFinv, however at present further work is needed to overcome design and technical issue encountered in this project.