Retrospective analysis evaluates CT and MRI fat quantification for diagnosing metabolic dysfunction–associated steatotic liver disease.

IntroductionEarly quantitative diagnosis of metabolic dysfunction–associated steatotic liver disease (MASLD) is crucial for preventing disease progression. Fat analysis and calculation technique (FACT) MRI fat quantification has shown potential diagnostic value. This study aimed to compare the diagnostic performance of CT and MRI fat quantification techniques and to evaluate their correlation with clinical and biochemical markers in in patients with MASLD.MethodsA retrospective analysis was conducted in 240 MASLD patients and 89 controls from December 2021 to February 2025. Demographic and clinical data, serum markers of liver function, inflammation, and lipid metabolism, and CT/MRI fat quantitative parameters were compared between groups. Correlations between CT/MRI parameters and serum markers were analyzed. All 329 participants were randomly divided into a training cohort (n = 230) and a validation cohort (n = 99) for diagnostic performance evaluation. Diagnostic performance was evaluated using ROC curves and AUC values, with the DeLong tests used for pairwise comparisons.ResultsMASLD patients showed higher levels of transaminases, glutamyl transpeptidase, interleukins, total cholesterol, and triglycerides (all P < 0.05). MRI-derived fat fraction (FF) and R2*, as well as CT-derived fat concentration and the liver-to-spleen CT ratio, were significantly elevated (P < 0.05). FF showed the highest diagnostic performance (AUC), followed by R2*. CT parameters demonstrated acceptable but lower diagnostic performance diagnostic performance than MRI parameters.ConclusionGiven the exploratory nature of the study and the lack of uniform gold standard validation, the diagnostic performance findings of FACT-derived fat parameters (FF and R2*) should be interpreted with caution. FACT-derived fat parameters showed significant correlations with CT-based measures and serum markers and demonstrated higher diagnostic performance compared to CT-based parameters. These findings suggest FACT-derived MRI parameters may serve as useful non-invasive tools for assessing hepatic fat content in MASLD. However, further validation with accurate reference standards, such as MRI-PDFF or liver biopsy, is required to confirm their clinical utility.