A single-center, prospective study included 178 participants with known or suspected liver disease. PCCT was performed using a dual-source system, and hepatic fat fraction was quantified via dedicated post-processing software. All examinations were acquired at 120 kVp in a portal venous phase. Comparative data were obtained using biopsy, CAP, or MRI PDFF (3T clinical system), with assessments stratified by steatosis grades. Patients included in the study had received one of the following comparative modalities for estimation of hepatic fat fraction within a 120-day window. PDFF were derived from six-echo, three-dimensional gradient-echo sequences (6E-mDixon) with equidistant echo spacing of 1.15 ms.

Fat fraction measurements were performed by placing one circular region of interest (ROI) with at least a 1-cm radius in each of the four defined liver segments (III, IVa, V and VII), avoiding major vessels, bile ducts, liver borders and artifacts (Figure 1). For histopathologic analysis steatosis was graded according to the Nonalcoholic Steatohepatitis Clinical Research Network grading system: 0 (<5% hepatocytes), 1 (5-33% hepatocytes), 2 (33-66% hepatocytes) and 3 (>66% hepatocytes). For US based assessement of steatosis four grades were differentiated:  0   (<238 dB/m), 1   (238-260 dB/m), 2   (260-290 dB/m) and 3 (>290 dB/m).

Statistical analysis included ICC, Spearman correlation, and ROC analysis.