Accurate and spatially stable quantification of hepatic steatosis using ultrasound-derived fat fraction in metabolic dysfunction-associated steatotic liver disease (MASLD): a prospective magnetic resonance imaging-proton density fat fraction (MRI-PDFF) study

Introduction

Metabolic dysfunction-associated steatotic liver disease (MASLD) is currently the most prevalent form of chronic liver disease worldwide, with a continuously increasing prevalence over the past two decades (1,2). Epidemiological data indicate that the prevalence of MASLD in China has risen from approximately 15% in the early 2000s to nearly 30%, making it one of the leading causes of chronic liver disease (3,4).

MASLD encompasses a continuous disease spectrum ranging from simple hepatic steatosis to metabolic dysfunction-associated steatohepatitis and may further progress to liver fibrosis, cirrhosis, and hepatocellular carcinoma (HCC) (5,6). Given its close association with obesity (7), metabolic syndrome (8), and insulin resistance (9,10), the risk of disease progression exhibits marked interindividual variability. Consequently, accurate and reproducible quantitative assessment of hepatic steatosis represents a central requirement for clinical risk stratification and treatment response evaluation in MASLD.

Liver biopsy remains the histopathological reference standard for assessing hepatic steatosis (11), however, its invasiveness, sampling variability, and limited reproducibility restrict its suitability for long-term follow-up and large-scale population screening. Although serum-based biomarkers have some screening utility, they remain inadequate for precise quantification of hepatic fat content (12). Magnetic resonance imaging-proton density fat fraction (MRI-PDFF) is widely regarded as the current noninvasive reference method for quantitative assessment of liver fat, yet its high cost and limited accessibility constrain its routine clinical use (13).

Ultrasonography plays an important role in MASLD assessment owing to its wide availability and repeatability (14). The controlled attenuation parameter is one of the most commonly used quantitative ultrasound metrics; however, its measurements are susceptible to body mass index (BMI) and other technical factors, with reduced diagnostic performance particularly in obese individuals (15,16). In recent years, ultrasound-derived fat fraction (UDFF) has emerged as a two-dimensional quantitative ultrasound technique that integrates attenuation and backscatter-related parameters to estimate hepatic fat content as a continuous percentage (17,18).

While several preliminary studies have demonstrated a good overall correlation between UDFF and MRI-PDFF (19-22), most existing investigations have primarily focused on global liver measurements or single-site sampling. However, hepatic steatosis is increasingly recognized as a spatially heterogeneous process, influenced by regional portal perfusion, intrahepatic vascular architecture, and technical factors such as acoustic window variability. Whether UDFF measurements obtained from different hepatic segments can reliably and consistently represent diffuse liver fat burden in patients with MASLD remains insufficiently explored.

Therefore, the present study aimed not only to evaluate the diagnostic accuracy of UDFF against MRI-PDFF, but also to investigate its spatial stability, reproducibility, and clinical correlates across the full spectrum of hepatic fat content. The inclusion of a healthy control group allowed assessment of discriminative performance across a wider biological range. We present this article in accordance with the STARD reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0236/rc).

Methods

Study population and study design

This prospective study consecutively enrolled 109 patients with MASLD who presented to Beijing Friendship Hospital, Capital Medical University, between September 2023 and January 2024. In addition, 20 healthy volunteers without known liver disease were recruited as the control group (Figure 1).

Figure 1 Graphical Abstract of the prospective MRI-PDFF-referenced study. Upper panel: study design involving 109 MASLD patients and 20 healthy volunteers. Lower panel: key findings showing that UDFF provides accurate liver fat quantification with high inter-observer reliability (ICC =0.98) and excellent discriminative power across all steatosis grades, independent of anthropometric metabolic factors. ALT, alanine aminotransferase; AST, aspartate aminotransferase; AUC, area under the curve; BMI, body mass index; GGT, gamma-glutamyl transferase; ICC, intraclass correlation coefficient; LDL-C, low-density lipoprotein cholesterol; MASLD, metabolic dysfunction-associated steatotic liver disease; MRI-PDFF, magnetic resonance imaging-proton density fat fraction; UDFF, ultrasound-derived fat fraction.

The diagnosis of MASLD was established according to the 2023 multisociety Delphi consensus statement. All patients with MASLD were required to have imaging-confirmed hepatic steatosis (by MRI-PDFF or ultrasonography) in combination with at least one cardiometabolic risk factor, including overweight or obesity (Asian population BMI ≥23 kg/m²), impaired fasting glucose or diabetes mellitus, hypertension (systolic or diastolic blood pressure >130/85 mmHg or current antihypertensive treatment), hypertriglyceridemia (≥1.70 mmol/L), or low high-density lipoprotein cholesterol (HDL-C; ≤1.0 mmol/L in men and ≤1.3 mmol/L in women).

Healthy controls were eligible if they were aged 18–65 years, had a BMI of 18.5–23 kg/m², and had no evidence of acute or chronic disease, metabolic syndrome, hepatic parenchymal disease, or abnormal liver function based on clinical evaluation, laboratory tests, and routine abdominal ultrasonography.

Exclusion criteria for both groups included liver injury due to other definite etiologies (viral hepatitis, autoimmune liver disease, inherited metabolic liver disease, or drug-induced liver injury); excessive alcohol consumption (≥210 g/week for men or ≥140 g/week for women within the previous 12 months); severe systemic diseases (cardiovascular, respiratory, renal, hematologic, endocrine, or psychiatric disorders); and pregnancy or lactation.

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ethics Committee of Beijing Friendship Hospital, Capital Medical University (No. 2023-P2-247-01). All participants were enrolled voluntarily and provided written informed consent.

Clinical data collection and metabolic index calculation

Demographic and anthropometric data, including age, sex, height, and body weight, were collected for all participants. BMI was calculated as weight (kg) divided by height squared (m²).

All blood samples were obtained after a fasting period of at least 8 hours and analyzed using an automated biochemical analyzer. Liver-related parameters included alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), and gamma-glutamyl transferase (GGT). Lipid metabolic parameters included total cholesterol (CHOL), triglycerides (TG), HDL-C, and low-density lipoprotein cholesterol (LDL-C).

The hepatic steatosis index (HSI), atherogenic index of plasma, TG/HDL-C ratio, and ALT/AST ratio were subsequently calculated. HSI was derived from a composite formula incorporating the ALT/AST ratio and BMI, whereas atherogenic index of plasma was defined as the common logarithm of the TG-to-HDL-C ratio.

UDFF measurement

Quantitative ultrasound assessment of hepatic fat content was performed using a Siemens ACUSON Sequoia ultrasound system equipped with a deep abdominal transducer (DAX) convex abdominal transducer (frequency range, 1.0–3.5 MHz). All UDFF measurements were acquired strictly in accordance with the manufacturer’s recommended acquisition protocol, without modification of the predefined algorithm parameters. UDFF is derived from a combined analysis of the ultrasound attenuation coefficient (AC) and backscatter coefficient (BSC) and is automatically calculated by a built-in algorithm to yield the percentage of hepatic fat content.

All ultrasound examinations were conducted after fasting for at least 8 hours. Scans were performed by two radiologists with more than 5 years of experience in liver ultrasonography, following a standardized protocol with identical transducers and imaging settings. Participants were examined in the supine position with the right arm elevated, and measurements were obtained in the right hepatic lobe to minimize interference from cardiac motion and gastrointestinal gas (Figure S1A-S1C).

A fixed ROI of 1.5×1.5 cm was utilized to balance signal averaging with the avoidance of macrovasculature, adhering to established quantitative ultrasound protocols. To minimize cardiac motion artifacts and ensure acoustic window stability, measurements were restricted to the right hepatic lobe (Couinaud segments V, VI, VII, and VIII). The ROI was consistently positioned 1.5–2.0 cm below the liver capsule, ensuring the exclusion of large vessels and bile ducts. UDFF values were recorded for each segment and averaged to yield UDFF mean as the representative metric for hepatic fat content. To ensure objectivity, sonographers remained blinded to participants’ clinical data and MRI-PDFF results during acquisition. To assess measurement reliability, approximately 30% of participants were randomly selected for repeated measurements performed independently by two sonographers, both blinded to each other’s results and to clinical information. Intraobserver and interobserver agreement of UDFF measurements were subsequently analyzed.

MRI-PDFF as the reference standard

All patients with MASLD underwent MRI-PDFF examination within one week after ultrasonography, which served as the reference standard for quantitative assessment of hepatic fat content. No therapeutic interventions that could influence liver fat content were administered between the two examinations. Based on MRI-PDFF values, hepatic steatosis was categorized into four grades: S0 (<5.2%), S1 (5.2–11.3%), S2 (11.3–17.1%), and S3 (>17.1%) (23).

Statistical analysis

All statistical analyses were performed using SPSS software (version 27.0; IBM Corp., Armonk, NY, USA) and R software (version 4.4.3). Normality of continuous variables was assessed using the Shapiro-Wilk test. Normally distributed data are presented as mean ± standard deviation, whereas non-normally distributed data are expressed as median (interquartile range). Comparisons among multiple groups were conducted using the Kruskal-Wallis H test.

Pearson or Spearman correlation analyses were used to evaluate associations between UDFF mean and metabolic parameters. Univariate and multivariable linear regression models were applied to identify independent clinical determinants of UDFF, with results reported as standardized β coefficients and 95% confidence intervals.

The diagnostic performance of UDFF at different steatosis thresholds was assessed using receiver operating characteristic (ROC) curve analysis, and optimal cut-off values were determined based on the Youden index. These cut-offs were derived from the study population using the Youden index and were not pre-specified by the manufacturer. To date, no universally standardized manufacturer-recommended cut-off has been established. Differences in the area under the curve (AUC) between UDFF and other indicators were compared using the DeLong test.

To evaluate the stability and reliability of UDFF across different right hepatic segments, the intraclass correlation coefficient (ICC) (24), concordance correlation coefficient (25), and coefficient of variation (CV) (26) were calculated. Agreement between UDFF and MRI-PDFF was assessed using the Bland-Altman method (27), with estimation of mean bias and 95% limits of agreement. All statistical tests were two-sided, and a P value <0.05 was considered statistically significant.

Results

Baseline characteristics and clinical-metabolic features across steatosis grades

A total of 129 participants were included and stratified according to MRI-PDFF into a normal group (n=20), mild steatosis group (n=13), moderate steatosis group (n=35), and severe steatosis group (n=61) (Table 1). Body weight and BMI increased progressively with steatosis severity (both P<0.001). Among liver enzymes, ALT, AST, and GGT differed significantly across steatosis grades (all P<0.001), whereas ALP showed no significant difference.

Regarding lipid metabolism, total cholesterol and LDL-C differed significantly across steatosis grades (P=0.013 and P=0.017, respectively), while TG, HDL-C, atherogenic index of plasma, and the TG/HDL-C ratio did not. Among composite metabolic indices, HSI increased in parallel with steatosis severity (P<0.001). Sex distribution differed across grades, whereas height was comparable among groups.

The distributions of representative clinical and biochemical parameters across MRI-PDFF-defined steatosis grades are illustrated in Figure 2A-2F. BMI (Figure 2A) and HSI (Figure 2D) increased significantly from S0 to S1 (P<0.001) but showed a plateau pattern at higher grades (S2–S3). GGT (Figure 2B) and the ALT/AST ratio (Figure 2E) exhibited a progressive increase, with the most pronounced change in ALT/AST observed between S1 and S2 (P<0.001). LDL-C (Figure 2C) differed only between S0 and S1 and demonstrated limited discrimination at higher grades.

Figure 2 Comparative distribution of clinical and ultrasonic parameters across MRI-PDFF-determined hepatic steatosis grades. Raincloud plots illustrate the distribution of each parameter according to MRI-PDFF grades (S0–S3). Boxplots indicate the median and IQR; half-violin plots show the density distribution; and jittered points represent individual observations. (A-E) The distributions of BMI, GGT, LDL-C, HSI score, and ALT/AST ratio across grades, respectively. (F) The distinct stepwise increase in UDFF mean values relative to steatosis severity. The P values from the Kruskal-Wallis test are indicated in the top-left of each panel. Horizontal lines represent pairwise comparisons between groups. Significance levels: ****, P<0.0001; ***, P<0.001; *, P<0.05. ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; GGT, gamma-glutamyl transferase; HSI, hepatic steatosis index; IQR, interquartile range; LDL-C, low-density lipoprotein cholesterol; MRI-PDFF, magnetic resonance imaging-proton density fat fraction; ns, not significant; UDFF, ultrasound-derived fat fraction.

In contrast, UDFF mean (Figure 2F) demonstrated the strongest association with MRI-PDFF grading, showing a clear and continuous stepwise increase from S0 to S3 (P<0.0001), with minimal overlap between adjacent grades. This pattern indicates a markedly higher discriminatory resolution of UDFF for steatosis severity compared with conventional biochemical markers and clinical indices.

Distribution and reliability of UDFF measurements across hepatic segments

The agreement and reproducibility of UDFF measurements obtained from different right hepatic segments (segments V–VIII) relative to MRI-PDFF were evaluated (Table S1). Reliability analyses demonstrated excellent consistency across all segments, with both ICC and concordance correlation coefficient values reaching 0.98. Bland-Altman analyses showed minimal mean bias (0.26–1.14%) and narrow 95% limits of agreement. CVs remained low across segments (9.25–10.16%), indicating good technical stability.

Segmental UDFF values are summarized in Table S2. UDFF measurements in all segments were significantly higher in patients with MASLD than in healthy controls (all P<0.001). Within the MASLD group, median UDFF values were comparable across segments. The averaged UDFF (UDFF mean) was 19.03% (14.03–25.00%) in the MASLD group and 2.06% (1.78–2.55%) in the control group.

Diagnostic performance of UDFF and clinical-metabolic indicators for steatosis grading

With increasing MRI-PDFF-defined steatosis grades, all evaluated clinical, biochemical, and ultrasound-derived parameters showed varying degrees of distributional expansion. In composite comparisons (Figure S2A), UDFF mean and HSI demonstrated the most pronounced stepwise separation across grades, whereas BMI and LDL-C showed substantial overlap at moderate-to-severe stages (S2–S3). Segmental consistency analysis further showed highly symmetric UDFF distributions across segments V–VIII within each steatosis grade (Figure S2B), indicating minimal spatial variability.

Using MRI-PDFF as the reference standard, UDFF demonstrated excellent diagnostic performance across all steatosis thresholds (Table 2). For detecting any steatosis (≥S1), UDFF mean achieved an AUC of 1.00 [95% confidence interval (CI): 1.00–1.00], with an optimal cut-off of 5.09%, yielding 100% sensitivity and specificity. HSI (AUC =0.97) and BMI (AUC =0.95) also showed good performance but were significantly inferior to UDFF mean (Figure 3A). For identifying moderate-or-greater steatosis (≥S2), UDFF mean achieved an AUC of 0.99 (95% CI: 0.99–1.00), with an optimal cut-off of 11.18%, sensitivity of 1.00, and specificity of 0.96, whereas the AUCs of the ALT/AST ratio and HSI declined to 0.88 and 0.86, respectively (Figure 3B). In the diagnosis of severe steatosis (≥S3), UDFF mean maintained an AUC of 0.99 (95% CI: 0.99–1.00), with an optimal cut-off of 18.46%, sensitivity of 0.93, and specificity of 0.96. In contrast, all clinical metabolic indicators showed AUCs below 0.70 (Figure 3C). DeLong tests confirmed that UDFF mean outperformed BMI, GGT, LDL-C, HSI, and the ALT/AST ratio across all thresholds (most P<0.001) (Table S3).

Figure 3 Comparison of the diagnostic performance of UDFF and clinical/biochemical indicators for identifying different grades of hepatic steatosis. ROC curves are shown for the identification of (A) steatosis grade ≥S1, (B) steatosis grade ≥S2, and (C) steatosis grade S3, with MRI-PDFF as the reference standard. The diagnostic models include UDFF mean, HSI score, BMI, ALT/AST ratio, GGT, and LDL-C. In all categories, UDFF mean demonstrates superior diagnostic performance compared to other clinical markers, with AUC values of 1.00 (95% CI: 1.00–1.00) for ≥S1, 0.99 (95% CI: 0.96–1.00) for ≥S2, and 0.99 (95% CI: 0.93–0.98) for S3. The diagonal dashed line represents the line of no discrimination (AUC =0.5). ALT, alanine aminotransferase; AST, aspartate aminotransferase; AUC, area under the curve; BMI, body mass index; CI, confidence interval; GGT, gamma-glutamyl transferase; HSI, hepatic steatosis index; LDL-C, low-density lipoprotein cholesterol; MRI-PDFF, magnetic resonance imaging-proton density fat fraction; ROC, receiver operating characteristic; UDFF, ultrasound-derived fat fraction.

Segment-specific analyses showed that UDFF mean and individual segmental UDFF values (segments V–VIII) consistently achieved AUCs ranging from 0.98 to 1.00, with comparable cut-off values, sensitivities, and specificities (Table S4), supporting robust intersegmental consistency.

Clinical correlates of UDFF

In univariate linear regression analyses, UDFF was significantly associated with ALT, AST, GGT, ALP, body weight, LDL-C, and age (Table 3). Positive associations were observed for ALT (β=0.41, P<0.001), GGT (β=0.39, P<0.001), AST (β=0.29, P=0.001), LDL-C (β=0.21, P=0.021), and body weight (β=0.19, P=0.036), whereas age showed a negative association (β=−0.25, P=0.006).

In multivariable regression analysis, after adjustment for potential confounders, ALT (β=0.29, P=0.033) and GGT (β=0.23, P=0.015) remained independently and positively associated with UDFF. Sex (β=−0.63, P=0.031) and age (β=−0.23, P=0.008) were independently and negatively associated with UDFF, whereas BMI and body weight were no longer statistically significant (Figure S3).

Correlation analyses supported these findings (Figure S4). UDFF mean showed positive correlations with all evaluated clinical and biochemical parameters, with the strongest correlations observed for GGT (R=0.39) and HSI (R=0.33) (both P<0.001).

Reproducibility and agreement of UDFF measurements

Interobserver and intraobserver reproducibility analyses demonstrated excellent reliability of UDFF measurements. Bland-Altman analysis showed a minimal mean interobserver bias of −0.15, with most values within the 95% limits of agreement (−3.67 to 3.37), and an ICC of 0.988 (Figure S5A). Intraobserver repeated measurements also showed excellent correlation (R=0.99) and reliability (ICC =0.988) (Figure S5B), indicating minimal operator dependence.

Discussion

B-mode ultrasonography is currently the most commonly used imaging modality for the assessment of hepatic steatosis (28). In this study, we found that UDFF was highly correlated with steatosis grades determined by MRI-PDFF, and exhibited a clear, stepwise increase with disease severity. These results suggest that UDFF holds considerable promise for quantitative diagnosis of hepatic steatosis in clinical practice. Prior to broader implementation, however, the repeatability and reliability of this technique must be carefully evaluated. Our study demonstrated excellent measurement stability of UDFF in adult liver assessment, with both inter‑observer and intraobserver ICCs reaching 0.988. This performance surpasses previously reported ICCs for single AC (29,30) or BSC (31) measurements (0.80–0.90), indicating that integrating these parameters into the composite UDFF metric can effectively suppress tissue scattering noise and reduce random errors introduced by operator variability, thereby providing a reliable benchmark for clinical follow‑up.

Given the high consistency of UDFF, we further investigated its spatial stability across different hepatic segments. Hemodynamic differences in portal vein perfusion can potentially result in spatial heterogeneity of hepatic fat infiltration (32), and previous studies have suggested that the right hepatic segment VIII exhibits the highest fat content (33). However, by comparing measurement agreement across hepatic segments V, VI, VII, and VIII at different steatosis grades, we found that UDFF measurements across right hepatic segments demonstrated excellent intersegmental agreement. However, measurements were limited to the right lobe in accordance with manufacturer recommendations, and spatial variability in the left hepatic lobe was not assessed. The ICCs for all segments consistently reached 0.98, indicating a high level of measurement consistency. Regarding measurement stability, CV analysis showed that UDFF values across all segments exhibited low variability, with CVs below 11%, suggesting limited overall measurement fluctuation. Nevertheless, mild intersegmental differences were observed. Specifically, segment VI showed relatively lower CV values, whereas segment VII exhibited slightly higher variability, indicating modest differences in random measurement fluctuation among segments during repeated acquisitions (34). Such variability may be attributable to technical and anatomical factors, including acoustic window quality, rib shadowing, and intrahepatic vascular distribution (34-36).

Conventional B-mode ultrasound assessment of steatosis was not systematically graded in this study. Therefore, we were unable to directly compare the diagnostic performance of UDFF with qualitative B-mode evaluation, particularly in mild steatosis, where conventional ultrasound is known to have reduced sensitivity. Future studies comparing UDFF with standardized B-mode grading systems in early steatosis are warranted.

We further analyzed clinical determinants influencing UDFF values. Univariate analysis revealed significant associations between UDFF and BMI (37), ALT (22), GGT (38), and LDL (39). However, multivariate regression identified age, ALT, and GGT as independent predictors, while BMI lost statistical significance (40). Notably, although BMI is a widely recognized indicator of obesity, its diagnostic performance markedly declined with increasing steatosis severity; for identifying severe steatosis (S3), BMI yielded an AUC of only 0.58, providing minimal clinical discrimination, whereas UDFF maintained a high AUC of 0.99. This highlights a critical biological distinction: peripheral obesity metrics predominantly reflect systemic energy imbalance, whereas UDFF directly reflects the biophysical accumulation of large lipid droplets within hepatocytes. Consequently, UDFF can accurately identify “lean hepatic steatosis” patients whose BMI may be normal yet who exhibit substantial intrahepatic fat deposition.

To validate diagnostic thresholds and agreement with the “gold standard”, Bland‑Altman analysis was performed. The mean interobserver bias was only −0.15, indicating excellent clinical agreement. Moreover, DeLong test results confirmed that UDFF mean outperformed conventional biochemical models such as the HSI, ALT/AST ratio, and GGT across all steatosis grades. Specifically, a UDFF cutoff of 5.09% achieved both sensitivity and specificity of 1.00 for detecting ≥S1 steatosis. It should be noted that the extremely high AUC values observed in this study, particularly for detecting ≥S1 steatosis, may partially reflect spectrum bias. The relatively clear separation between MASLD patients and healthy controls, along with the imbalance in sample size (109 MASLD vs. 20 controls), may have amplified diagnostic discrimination. Therefore, the reported diagnostic performance should be interpreted with caution and validated in more heterogeneous populations. These findings have important implications for clinical workflow: UDFF can provide MRI‑comparable quantitative accuracy during outpatient screening, potentially reducing the need for expensive and time‑consuming MRI examinations.

While the findings are promising, several limitations should be acknowledged. First, this was a single-center study conducted using a single ultrasound platform, and external validation across different vendors, operators, and populations is required to confirm generalizability. Second, although MRI-PDFF is widely accepted for quantitative fat assessment, histopathological confirmation was not available, limiting direct correlation with microscopic steatosis grading. In addition, coexisting pathological processes such as fibrosis or inflammatory activity were not systematically characterized. Elastography techniques (2D shear wave elastography or point shear wave elastography) were not performed to exclude significant fibrosis, and advanced fibrosis may alter hepatic acoustic properties, potentially influencing UDFF measurements. Furthermore, individuals with extreme BMI values were underrepresented, and severe obesity may reduce acoustic penetration and affect quantitative ultrasound performance. Finally, only right-lobe segments were evaluated in accordance with current manufacturer guidance; left-lobe variability, which may exhibit greater heterogeneity, was not assessed and warrants future investigation.

Conclusions

In summary, this study demonstrates that UDFF, as an ultrasound‑based quantitative technique, provides excellent diagnostic performance and reproducibility for the assessment of hepatic steatosis in patients with MASLD, showing high concordance with MRI‑PDFF. Its ease of operation and high repeatability suggest considerable clinical utility. Future multicenter, longitudinal studies are warranted to further clarify the value of UDFF in disease monitoring and therapeutic evaluation.

Acknowledgments

The authors would like to thank all participants for their involvement in this study. The authors also acknowledge the clinical and technical staff of the Department of Ultrasound and the Department of Radiology at Beijing Friendship Hospital, Capital Medical University, for their assistance in data acquisition and logistical support.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0236/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0236/dss

Funding: This study was supported by the Joint funds of Capital’s Funds for Health Improvement and Research (Grant No. SFMR-2026-M2-2-111-1). The funding source had no role in the study design; data collection, analysis, or interpretation; manuscript preparation; or the decision to submit the article for publication.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0236/coif). S.R.W., J.F.Z. and X.D.H. report support from the Joint funds of Capital’s Funds for Health Improvement and Research. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ethics Committee of Beijing Friendship Hospital, Capital Medical University (No. 2023-P2-247-01). All participants were enrolled voluntarily and provided written informed consent.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Devasia AG, Ramasamy A, Leo CH. Current Therapeutic Landscape for Metabolic Dysfunction-Associated Steatohepatitis. Int J Mol Sci 2025;26:1778. [Crossref] [PubMed]
2. Pîrvu BF, Clenciu D, Beldie LA, Dica CC, Burticală MA, Ţenea-Cojan TŞ, Mitrea A, Amzolini AM, Efrem IC, Mogoş GFR, Vladu IM. The burden of cancer in metabolic dysfunction-associated steatotic liver disease. Rom J Morphol Embryol 2024;65:627-35. [Crossref] [PubMed]
3. Li R, Li H, Ye X, Qin J. Metabolic Dysfunction-associated Steatotic Liver Disease is Becoming the Leading Driver of the Burden of Cirrhosis in China: Results From the Global Burden of Disease Study 2019. J Clin Gastroenterol 2025;59:592-7. [Crossref] [PubMed]
4. Jain P, Jain A, Deshmukh R, Samal P, Satapathy T. Ajazuddin. Metabolic dysfunction-associated steatotic liver disease (MASLD): Exploring systemic impacts and innovative therapies. Clin Res Hepatol Gastroenterol 2025;49:102584. [Crossref] [PubMed]
5. Alpízar Salazar M, Olguín Reyes SE, Medina Estévez A, Saturno Lobos JA, De Aldecoa Castillo JM, Carrera Aguas JC, Alaniz Monreal S, Navarro Rodríguez JA, Alpízar Sánchez DMF. Natural History of Metabolic Dysfunction-Associated Steatotic Liver Disease: From Metabolic Syndrome to Hepatocellular Carcinoma. Medicina (Kaunas) 2025;61:88. [Crossref] [PubMed]
6. Saito T, Tsuchishima M, Tsutsumi M, George J. Molecular pathogenesis of metabolic dysfunction-associated steatotic liver disease, steatohepatitis, hepatic fibrosis and liver cirrhosis. J Cell Mol Med 2024;28:e18491. [Crossref] [PubMed]
7. Jiang Z, Wang Y, Zhao X, Cui H, Han M, Ren X, Gang X, Wang G. Obesity and chronic kidney disease. Am J Physiol Endocrinol Metab 2023;324:E24-E41. [Crossref] [PubMed]
8. Zhou M, Sun W, Gao Y, Jiang B, Sun T, Xu R, Zhang X, Wang Q, Xuan Q, Ma S. Metabolomic profiling reveals interindividual metabolic variability and its association with cardiovascular-kidney-metabolic syndrome risk. Cardiovasc Diabetol 2025;24:315. [Crossref] [PubMed]
9. Caputo T, Gilardi F, Desvergne B. From chronic overnutrition to metaflammation and insulin resistance: adipose tissue and liver contributions. FEBS Lett 2017;591:3061-88. [Crossref] [PubMed]
10. Elfaki I, Mir R, Elnageeb ME, Hamadi A, Alharbi ZM, Bedaiwi RI, Javid J, Alrasheed T, Alatawi D, Alrohaf BM, Abunab MK, Muqri TA. Identification of Interactive Genetic Loci Linked to Insulin Resistance in Metabolic Syndrome-An Update. Medicina (Kaunas) 2025;61:83. [Crossref] [PubMed]
11. Kazi IN, Kuo L, Tsai E. Noninvasive Methods for Assessing Liver Fibrosis and Steatosis. Gastroenterol Hepatol (N Y) 2024;20:21-9.
12. Tamaki N, Ajmera V, Loomba R. Non-invasive methods for imaging hepatic steatosis and their clinical importance in NAFLD. Nat Rev Endocrinol 2022;18:55-66. [Crossref] [PubMed]
13. Starekova J, Hernando D, Pickhardt PJ, Reeder SB. Quantification of Liver Fat Content with CT and MRI: State of the Art. Radiology 2021;301:250-62. [Crossref] [PubMed]
14. Seneviratne N, Fang C, Sidhu PS. Ultrasound-based hepatic fat quantification: current status and future directions. Clin Radiol 2023;78:187-200. [Crossref] [PubMed]
15. Tamaki N, Kurosaki M, Yasui Y, Tsuchiya K, Izumi N. Attenuation coefficient (ATT) measurement for liver fat quantification in chronic liver disease. J Med Ultrason (2001) 2021;48:481-7. [Crossref] [PubMed]
16. Ren X, Xia S, Zhang L, Li R, Zhou W, Ji R, Zhou J, Tian J, Zhan W. Analysis of liver steatosis analysis and controlled attenuation parameter for grading liver steatosis in patients with chronic hepatitis B. Quant Imaging Med Surg 2021;11:571-8. [Crossref] [PubMed]
17. Huang Y, Li J, Liu C, Song D, Zhu C, Ren Y, et al. Noninvasive Quantification of Hepatic Steatosis Using Ultrasound-Derived Fat Fraction (CHESS2303): A Prospective Multicenter Study. MedComm (2020) 2025;6:e70123.
18. Niu KX, Ye JJ, Liu XQ, Li TG, Ma B, Yang QQ. Trimester-specific reference values of ultrasound-derived fat fraction and automated point shear-wave elastography for liver assessment in healthy singleton pregnancies. Quant Imaging Med Surg 2026;16:81. [Crossref] [PubMed]
19. Tavaglione F, Flagiello V, Terracciani F, Gallo P, Capparelli E, Spiezia C, De Vincentis A, Palermo A, Scriccia S, Galati G, Napoli N, Daniels SJ, Blau JE, Carlsson B, Khazrai YM, Incalzi RA, Picardi A, Vespasiani-Gentilucci U. Non-invasive assessment of hepatic steatosis by ultrasound-derived fat fraction in individuals at high-risk for metabolic dysfunction-associated steatotic liver disease. Diabetes Metab Res Rev 2024;40:e3787. [Crossref] [PubMed]
20. Jeon SK, Lee JM. Inter-platform reproducibility of ultrasound-based fat fraction for evaluating hepatic steatosis in nonalcoholic fatty liver disease. Insights Imaging 2024;15:46. [Crossref] [PubMed]
21. Meng L, Wang J, Yang H, Hu Y, Yang Z. Ultrasound-derived fat fraction to assess liver steatosis in obese patients with polycystic ovary syndrome. Clin Exp Med 2025;25:130. [Crossref] [PubMed]
22. Liu Q, Lyu X, Ren H, Wang M, Liu Y, Tang R, Liu J. Assessment of hepatic steatosis in children with metabolic dysfunction-associated steatotic liver disease using ultrasound derived fat fraction: first large-scale clinical evaluation. Quant Imaging Med Surg 2025;15:5582-92. [Crossref] [PubMed]
23. Imajo K, Kessoku T, Honda Y, Tomeno W, Ogawa Y, Mawatari H, Fujita K, Yoneda M, Taguri M, Hyogo H, Sumida Y, Ono M, Eguchi Y, Inoue T, Yamanaka T, Wada K, Saito S, Nakajima A. Magnetic Resonance Imaging More Accurately Classifies Steatosis and Fibrosis in Patients With Nonalcoholic Fatty Liver Disease Than Transient Elastography. Gastroenterology 2016;150:626-637.e7. [Crossref] [PubMed]
24. Raykov T, Marcoulides GA. Intraclass Correlation Coefficients in Hierarchical Design Studies With Discrete Response Variables: A Note on a Direct Interval Estimation Procedure. Educ Psychol Meas 2015;75:1063-70. [Crossref] [PubMed]
25. Tsai MY, Sun CN, Lin CC. Concordance correlation coefficients estimated by modified variance components and generalized estimating equations for longitudinal overdispersed Poisson data. Stat Methods Med Res 2022;31:267-86. [Crossref] [PubMed]
26. Pélabon C, Hilde CH, Einum S, Gamelon M. On the use of the coefficient of variation to quantify and compare trait variation. Evol Lett 2020;4:180-8. [Crossref] [PubMed]
27. Doğan NÖ. Bland-Altman analysis: A paradigm to understand correlation and agreement. Turk J Emerg Med 2018;18:139-41. [Crossref] [PubMed]
28. Tan ZX, Mehta B, Kusel K, Seow J, Zelesco M, Abbott S, Simons R, Boardman G, Welman CJ, Ayonrinde OT. Hepatic steatosis: Qualitative and quantitative sonographic assessment in comparison to histology. Australas J Ultrasound Med 2024;27:179-88. [Crossref] [PubMed]
29. Dioguardi Burgio M, Ronot M, Reizine E, Rautou PE, Castera L, Paradis V, Garteiser P, Van Beers B, Vilgrain V. Quantification of hepatic steatosis with ultrasound: promising role of attenuation imaging coefficient in a biopsy-proven cohort. Eur Radiol 2020;30:2293-301. [Crossref] [PubMed]
30. Ferraioli G, Raimondi A, Maiocchi L, De Silvestri A, Poma G, Kumar V, Barr RG. Liver Fat Quantification With Ultrasound: Depth Dependence of Attenuation Coefficient. J Ultrasound Med 2023;42:2247-55. [Crossref] [PubMed]
31. Paige JS, Bernstein GS, Heba E, Costa EAC, Fereirra M, Wolfson T, Gamst AC, Valasek MA, Lin GY, Han A, Erdman JW Jr, O'Brien WD Jr, Andre MP, Loomba R, Sirlin CB. A Pilot Comparative Study of Quantitative Ultrasound, Conventional Ultrasound, and MRI for Predicting Histology-Determined Steatosis Grade in Adult Nonalcoholic Fatty Liver Disease. AJR Am J Roentgenol 2017;208:W168-77. [Crossref] [PubMed]
32. Borji S, Sani HS, Ghorbani S, Soltani M, Mohammadi V, Elahi R. The correlation between metabolic dysfunction-associated steatotic liver disease (MASLD) grades and hemodynamic alterations of the portal, hepatic, and splenic vein and spleen size. J Ultrasound 2025;28:339-47. [Crossref] [PubMed]
33. Song D, Wang P, Han J, Chen H, Gao R, Li L, Li J. Reproducibility of ultrasound-derived fat fraction in measuring hepatic steatosis. Insights Imaging 2024;15:254. [Crossref] [PubMed]
34. Fetzer DT, Vijay K, Caserta MP, Patterson-Lachowicz A, Dahiya N, Rodgers SK. Artifacts and Technical Considerations at Contrast-enhanced US. Radiographics 2023;43:e220093. [Crossref] [PubMed]
35. Agnew AM, Murach MM, Dominguez VM, Sreedhar A, Misicka E, Harden A, Bolte JH 4th, Kang YS, Stammen J, Moorhouse K. Sources of Variability in Structural Bending Response of Pediatric and Adult Human Ribs in Dynamic Frontal Impacts. Stapp Car Crash J 2018;62:119-92. [Crossref] [PubMed]
36. Yancey K, Ruddy L, Simmons CL, Crawford DA, Alzubaidi S, Naidu S, Ozen M. Unveiling Portal Vein Anatomy: Clinical Impact and Significance of Variants in Transjugular Intrahepatic Portosystemic Shunt Creation. Semin Intervent Radiol 2025;42:156-65. [Crossref] [PubMed]
37. Huang YL, Cheng J, Wang Y, Xu XL, Wang SW, Wei L, Dong Y. Hepatic steatosis using ultrasound-derived fat fraction: First technical and clinical evaluation. Clin Hemorheol Microcirc 2024;86:51-61. [Crossref] [PubMed]
38. Xing Y, Chen J, Liu J, Ma H. Associations Between GGT/HDL and MAFLD: A Cross-Sectional Study. Diabetes Metab Syndr Obes 2022;15:383-94. [Crossref] [PubMed]
39. Wang P, Song D, Han J, Zhang J, Chen H, Gao R, Shen H, Li J. Comparing Three Ultrasound-Based Techniques for Diagnosing and Grading Hepatic Steatosis in Metabolic Dysfunction-Associated Steatotic Liver Disease. Acad Radiol 2025;32:1949-57. [Crossref] [PubMed]
40. Yang L, Ge Y, Zhu Q, Zhang Q, Wang L, Wang X, Yang Y, Shi H. The Combination of Fat Distribution and BMI Redefines Obesity: Result From NHANES. J Cachexia Sarcopenia Muscle 2025;16:e70013. [Crossref] [PubMed]