Abstract: Objective The purpose of this study was to explore diagnostic efficacy of non-alcoholic fatty liver diseases (NAFLD) by quantitative CT (QCT) measurement in individuals for physical examination. Methods 935 individuals were encountered for physical examination in our hospital between July 2020 and March 2022, and all underwent ultrasonography (US) and CT scan. Computed tomography value and bone mineral concentration were measured at four regions of interest by QCT Pro Model 4 system, and liver fat contents were calculated automatically. Receiver operating characteristic curve was drawn and the area under the curve (AUC) was calculated for evaluating diagnostic performance. Results Of 935 persons in our series, ultrasonography found NAFLD in 264 cases(28.2%), while it was 104 cases(11.1%) by liver /spleen CT value ratio; liver fat content in 264 persons with NAFLD proven by US was 8.7(6.6, 14.8), much higher than [4.6(3.4, 6.3), Z=-18.019, P<0.001] in 671 persons without NAFLD; liver fat content in 104 individuals with NAFLD diagnosed by CT was 16.7(13.5, 21.1), much higher than [5.2(3.8, 6.9), Z=-16.349, P<0.001] in 831 individuals without NAFLD; the sensitivity (Se) was 69.3% and the specificity (Sp) was 86.7%, based on ultrasonography diagnosis, when liver fat contents greater than 3.6% was set as the cut-off-value; the Se was 93.3% and the Sp was 98.4%, based on liver/spleen CT value ratio diagnosis, when liver fat contents equal to 2.1% was set as the cut-off-value. Conclusion Diagnosis of fatty liver by QCT scan is objective with reliability, which might provide a choice for clinical application.

Key words: Non-alcoholic fatty liver diseases, Physical examination, Quantitative CT, Diagnosis