To identify potential risk factors for chronic kidney disease (CKD) and cardiovascular disease (CVD) in Saudi Arabian patients with non-alcoholic fatty liver disease (NAFLD).

A 6-year prospective cohort study was carried out from January 2018 to January 2023, enrolling 1,500 patients. Data were initially collected between January 2018-2019, and follow-up assessments were carried out annually from 2020-2023. Patients were evaluated using biomarker analyses, and medical records were reviewed to assess the incidence of CVD and CKD. Biomarker levels, including blood creatinine, urine albumin, glomerular filtration rate (GFR), atherogenic index of plasma (AIP), and other key parameters, were monitored throughout the study.

Of the 1500 patients, 735 (49%) were diagnosed with NAFLD in 2018, while 765 (51%) did not have the condition. Compared to non-NAFLD patients, those with NAFLD exhibited elevated blood creatinine levels, lower GFR, and higher urine albumin, a key marker of kidney damage that is strongly linked to accelerated CKD progression and increased cardiovascular risk. Furthermore, NAFLD patients had a significantly higher AIP. Over the 4-year follow-up, these individuals showed a notable rise in the incidence of CKD and CVD.

Non-alcoholic fatty liver disease is associated with alterations in lipid profiles, elevated high-sensitivity C-reactive protein (hs-CRP), and elevated aspartate aminotransferase (AST), all of which may contribute to early CKD and CVD development. Older, obese individuals with NAFLD, high triglyceride, AST, and hs-CRP levels are at an elevated risk for these diseases.

Metabolic dysfunction-associated steatotic liver disease (MASLD) typically presents as "macrovesicular steatosis", where each hepatocyte contains a large fat vacuole (30-50 µm), indicating a more indolent form. In about 20% of cases, "microvesicular steatosis" occurs, with smaller vacuoles (1-15 µm) linked to steatohepatitis, cirrhosis progression, and increased risk of liver cancer. Emerging quantitative ultrasound (QUS) liver fat quantification (QUS-LFQ) tools measure various acoustic properties, but few methods compare techniques and imaging modalities, and the impact of fat vacuole size remains unclear. This study introduces a methodology to create ultrasound (US) phantoms that replicate fat vesicle size in MASLD. While imaging phantoms validate quantitative tools, no model currently links QUS-LFQ measurements to steatosis severity. Existing homogeneous phantoms assessing properties like attenuation, backscatter, and speed of sound overlook the microstructure of steatosis, despite the known effect of particle size on acoustic interactions. Here, agar-based phantoms simulate fat accumulation in steatotic hepatocytes using stable peanut oil droplets as analogs for lipid vacuoles. Microscopy and sizing confirm stability at 4 °C, 23 °C, and 50 °C. Both microscopy and US imaging confirm uniform distribution, with QUS-LFQ measurements reflecting fat content. These phantoms hold promise for validating quantitative imaging methods, particularly for US-based MASLD screening tools.

Wilson's disease (WD) liver damage is caused by abnormal copper deposition in the liver. Whether the degree of hepatic impairment correlates with the severity of liver involvement on ultrasound imaging remains unclear. This study aimed to categorize the liver echo patterns of WD according to the B-ultrasound and correlated the degree of hepatic injury with different liver echo patterns based on Transient elastography (TE) and Sound Touch Viscoelastography (STVi). A retrospective study was conducted on 106 WD patients. A consistency analysis was performed on liver stiffness measurement (LSM) through TE and STVi. The liver echo characteristics of all patients were categorized through B-ultrasound, and the differences in LSMTE, LSMSTVi, demographic and anthropometric data, and serological indicators among different patterns were compared. Both TE and STVi measurements exhibited high consistency in intra-observer and inter-observer differences. There were 6 echo patterns in the liver parenchyma of WD, among which "block-like" fatty liver and nodule patterns were more characteristic. The 6 types of liver echo patterns were classified into three groups: fatty liver pattern, coarse pattern, and nodule pattern. There was no significant difference in age, gender, and body mass index among the three groups. The comparison results of LSMTE and LSMSTVi among the three groups showed that the coarse pattern was higher than the fatty liver pattern, and there was no significant difference between the nodule pattern and the coarse pattern or the fatty liver pattern. The P III N-P of fatty liver and coarse patterns was higher than that of the nodule pattern. Characteristic ultrasound features of liver parenchyma help us diagnose WD. The degree of liver involvement on imaging may not correspond consistently with the actual level of liver damage. STVi can be used as a reliability method to evaluate the degree of WD liver damage quantitatively, and its other application values in WD are worth further exploration.

To investigate the relationship between hepatic steatosis index (HSI) and endometrial cancer (EC) and its diagnostic value for EC.

A total of 114 patients with pathologically diagnosed EC in Mindong Hospital, Ningde City, Fujian Province from 2016 to 2022 were retrospectively included as the EC group. A total of 175 patients with pathologically confirmed benign endometrial lesions (endometrial polyps and uterine submucosal fibroids) in the same hospital during the same period were selected as the control group. Non-parametric test were used to compare the differences in HSI and non-alcoholic fatty liver disease (NAFLD) between the two groups, and the diagnostic value of HSI and NAFLD levels on EC was analysed. The cut-off point of continuous variables was determined by receiver operating characteristic (ROC) curve analysis. Logistic regression analysis was used to calculate odds ratios (ORs).

The results showed that compared with the control group, serum GGT, CA125, HDL-C and HSI were significantly increased in the EC group (P<0.05). 27.19% of the EC patients (31/114) and 12% of the control group (21/175) had NAFLD, and the difference between the two groups was statistically significant (P<0.05). The results of univariate logistic regression analysis showed that GGT, CA125, HDL-C, HSI and NAFLD were significantly correlated with the occurrence of EC (P<0.05). Further multivariate logistic regression analysis showed that CA125 and HSI elevation were independent risk factors for EC (P<0.05).

NAFLD is closely associated with EC, and elevated HSI is an independent risk factor for EC.

Chronic liver disease is a world-wide epidemic. Any etiology that causes inflammation in the liver will lead to chronic liver disease. Presently, the most common inciting factor worldwide is steatotic liver disease. Recent advances in ultrasound imaging provide a multiparametric ultrasound methodology of diagnosing, staging, and monitoring treatment of chronic liver disease. Elastography has become a standard of care technique for the evaluation of liver fibrosis. Quantitative ultrasound allows for determination of the degree of fatty infiltration of the liver. Portal hypertension is the most important factor in determination of liver decompensation. B-mode findings combined with Doppler, and elastography techniques provide qualitative and quantitative methods of determining clinically significant portal hypertension. A newer method using contrast enhanced ultrasound may allow for a non-invasive quantitative estimation of the portal pressures. This paper reviews the use of multiparametric ultrasound in the evaluation of chronic liver disease including conventional B-mode ultrasound, Doppler, elastography and quantitative ultrasound for estimation of liver fat. The recent guidelines are presented and advised protocols reviewed.

There is growing evidence linking growth differentiation factor 15 (GDF15) to both metabolic dysfunction-associated steatotic liver disease (MASLD) and cardiovascular (CV) risk. Nevertheless, the potential relationship between circulating levels of GDF15 and key features of MASLD being predisposed to atherosclerotic CV disease is not fully unveiled. The aim of this study was to deepen into the role of circulating GDF15 levels on metabolic-associated liver injury and atherosclerotic CV disease. We determined the serum GDF15 levels in 156 participants of a metabolic patient-based cohort, and cross-sectionally explored its associations with liver injury and an advanced atherosclerotic lipoprotein profile assessed by nuclear magnetic resonance (1H-NMR). Additionally, we prospectively evaluated the association between GDF15 levels at baseline and incident atherosclerotic CV disease after a 10-year follow-up. GDF15 was related to liver injury and inflammatory hallmarks, and it increased the likelihood for liver steatosis independently of confounding factors. Likewise, GDF15 was positively associated with an atherogenic profile, particularly with the number of very-low-density lipoproteins (VLDL) particles and its cholesterol and triglyceride content, and with an indicator of subclinical atherosclerosis (i.e., carotid intima-media thickness (cIMT)). The baseline serum GDF15 levels were higher in the patients with atherosclerotic CV disease (10.6%) after a 10-year follow-up than in the individuals without CV disease. Altogether, this study provides new insights into the role of GDF15 in both MASLD and CV disease.

Serum ferritin has been suggested as a potential biomarker associated with disease progression in metabolic dysfunction-associated steatotic liver disease (MASLD).

We investigated the association between serum ferritin levels and liver-related events (LREs) in individuals with steatotic liver disease (SLD).

This cohort study included 17,560 adults with SLD (MASLD [n = 15,744], MASLD with increased alcohol intake (MetALD) [n = 1103] and cryptogenic SLD [n = 713]) without LRE at baseline. A steatotic liver was diagnosed using ultrasound, and LRE was defined as the development of decompensation (ascites, variceal bleeding and hepatic encephalopathy) or hepatocellular carcinoma. Participants were categorised into high (≥ 300 μg/L for males, ≥ 200 μg/L for females) or normal to low (< 300 μg/L for males, < 200 μg/L for females) ferritin levels.

During 211,425 person-years of follow-up (median: 12.3 years), 74 incident LRE cases were identified, with 63 cases in MASLD, 10 in MetALD and 1 in cryptogenic SLD. The multivariable-adjusted hazard ratio (aHR) for LRE comparing individuals with high and normal-to-low ferritin level was 3.13 (95% confidence interval [CI] 1.89-5.18). Increased risk of LRE in individuals with high serum ferritin level compared to those with normal to low serum ferritin level was consistent across SLD subtypes (aHR 2.69, 95% CI 1.55-4.67 for MASLD; aHR 5.73, 95% CI 1.31-25.0 for MetALD), and SLD severity assessed by Fibrosis-4 (FIB-4) index (aHR 2.38, 95% CI 1.34-4.21 for FIB-4 ≥ 1.3; aHR 3.13, 95% CI 1.18-8.29 for FIB-4 < 1.3).

Serum ferritin levels correlated with the risk of LRE in patients with SLD.

To explore the connection between metabolic parameters and the severity of hepatic steatosis determined through ultrasound in elderly individuals with metabolic dysfunction-associated fatty liver disease (MAFLD).

4,663 senior individuals who were 65 years of age or older were included in this research. They were examined physically at the Ninghai Street Community Health Service Center in Yantai City between June 7, 2021, and October 15, 2021. There were two categories of individuals identified: the MAFLD group (n = 2,985) and the non-MAFLD group (n = 1,678). Based on liver ultrasonography results, individuals in the MAFLD group were further separated into three groups: mild (n = 2,104), moderate (n = 766), and severe (n = 115). To identify indicators of risk for the severity of hepatic steatosis, metabolic data was contrasted between the groups employing logistic regression.

In comparison to the non-MAFLD group, the MAFLD group showed significantly elevated levels of body mass index (BMI), blood pressure, gender, age, lipid profile, alanine transaminase (ALT), and fasting blood glucose (FBG; p < 0.05). Among individuals with MAFLD, there was a positive correlation between BMI, FBG, ALT, and aspartate transaminase (AST) levels and the severity of hepatic steatosis (p < 0.05). Logistic regression analysis indicated that BMI, female gender, FBG, ALT, triglycerides (TG), and serum uric acid (SUA) constituted risk factors for increased severity of hepatic steatosis in MAFLD.

The severity of hepatic steatosis in elderly MAFLD patients is significantly correlated with female gender, BMI, ALT, FBG, TG, and SUA.

Abdominal point-of-care ultrasound is an essential diagnostic tool for internal medicine physicians. It can identify intraperitoneal free fluid, evaluate the liver for size, presence of steatosis, and assessment for possible cirrhosis. Diagnosing cholelithiasis or cholecystitis can expedite care. Physical examination for mild splenomegaly may be insensitive. In such cases, sonographic measurements may provide a more definitive diagnosis. With the proper training, these organs can be evaluated at the bedside and guide clinical decision making.

There is an unmet need for a biomarker of liver fat. We identified dimethylguanidino valeric acid (DMGV) as a circulating biomarker of liver fat. Here, we assess its two isoforms-symmetric (SDGV) and asymmetric (ADGV)-as biomarkers of steatosis. We determined plasma ADGV, SDGV, related metabolites, alanine aminotransferase (ALT), and the fatty liver index (FLI) in two cohorts and compared their diagnostic performance for liver fat detection. SDGV was the strongest predictor of moderate to severe steatosis. Changes in SDGV correlated with changes in liver fat % in a prospective cohort. In a murine model of fatty liver disease, protein levels and activity of alanine:glyoxylate aminotransferase 2 (AGXT2), which produces SDGV, were increased and coincided with elevation of SDGV concentrations. SDGV is a biomarker of liver fat and its increase in hepatic steatosis results from the upregulation of AGXT2 activity.

The Fontan operation, which directly connects the superior and inferior vena cava to the pulmonary artery, is a palliative surgery for children with a functional or anatomic single ventricle. This procedure leads to hemodynamic changes (Fontan circulation) in patients, who tend to develop congestive hepatic fibrosis characterized by sinusoidal fibrosis and dilatation beginning approximately 10 years after the procedure. In addition, in the context of severe fibrosis and cirrhosis, hepato-gastrointestinal complications including hepatocellular carcinoma, focal nodular hyperplasia, and portal hypertension can arise. Fontan-associated liver disease (FALD) encompasses the broad spectrum of liver alterations secondary to postoperative hemodynamic changes, and the effective management of FALD requires contributions from specialists in hepatology, gastroenterology, surgery, radiology, histopathology, and pediatric and adult cardiology. In this article, we outline the pathogenesis of FALD and discuss the importance of a multidisciplinary collaborative approach to its management.

Modifiable risk factors associated with cognitive functioning are important for identifying potential targets for intervention development. Although there are a few recognized modifiable risk factors (e.g., diabetes mellitus, diet, physical activity), there are limitations in the conclusions that can be drawn due to limited data. Therefore, this study examined the relationship between modifiable liver disease-linked metabolic and behavioral factors in a sample of community dwelling adults who do not currently experience functional limitations due to cognitive abilities. Individuals aged 19 to 69 were recruited to participate in this cross-sectional study in the Washington, DC area. Participants were assessed using anthropometric measures, ultrasound of the liver, glycated hemoglobin A1C, self-reported fatigue, clinical history, and 7 domains of cognitive function: processing speed, short- and long-term visual memory, working memory, inhibition, shifting, and abstract reasoning. The study included 104 participants (44% female, 51.1 ± 13.5 years old). The modifiable factors that were most consistently related to cognitive performance were waist-to-height ratio, which was related to a decrease in performance in 4 of the domains (short-term and long-term visual memory, working memory, and abstract reasoning), and the presence of nonalcoholic fatty liver disease, which was related to an increase in performance in the same 4 domains. This study suggests that liver disease-linked modifiable factors are associated with cognitive performance, even in middle-aged individuals without self-reported cognitive dysfunction. Further research is needed to explore the mechanisms that impact cognitive performance in relation to these factors to establish early intervention targets for reducing future cognitive deficits.

Metabolic dysfunction-associated steatotic liver disease (MASLD) represents a significant global health burden without established curative therapies. Early detection and preventive strategies are crucial for effective MASLD management. This study aimed to develop and validate machine-learning (ML) algorithms for accurate MASLD screening in a geographically diverse, large-scale population.

Data from the prospective Fasa Cohort Study, initiated in rural Fars province, Iran (March 2014), were employed for this purpose. The required data were collected using blood tests, questionnaires, liver ultrasonography, and physical examinations. A two-step approach identified key predictors from over 100 variables: (1) statistical selection using mean decrease Gini in random forest and (2) incorporation of clinical expertise for alignment with known MASLD risk factors. The hold-out validation approach (with a 70/30 train/validation split) was utilized, along with 5-fold cross-validation on the validation set. Logistic regression, Naïve Bayes, support vector machine, and light gradient-boosting machine (LightGBM) algorithms were compared for model construction with the same input variables based on area under the receiver operating characteristic curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy.

A total of 6,180 adults (52.7% female) were included in the study, categorized into 4816 non-MASLD and 1364 MASLD cases with a mean age (±standard deviation [SD]) of 48.12 (±9.61) and 49.47 (±9.15) years, respectively. Logistic regression outperformed other ML algorithms, achieving an accuracy of 0.88 (95% confidence interval [CI]: 0.86-0.89) and an AUC of 0.92 (95% CI: 0.90-0.93). Among more than 100 variables, the key predictors included waist circumference, body mass index (BMI), hip circumference, wrist circumference, alanine aminotransferase levels, cholesterol, glucose, high-density lipoprotein, and blood pressure.

Integration of ML in MASLD management holds significant promise, particularly in resource-limited rural settings. Additionally, the relative importance assigned to each predictor, particularly prominent contributors such as waist circumference and BMI, offers valuable insights into MASLD prevention, diagnosis, and treatment strategies.

Steatotic liver disease (SLD) is a new overarching term proposed to replace nonalcoholic fatty liver disease and metabolic dysfunction-associated fatty liver disease. Subclassification includes metabolic dysfunction-associated SLD (MASLD), MASLD with increased alcohol intake (MetALD), and cryptogenic SLD. This study aimed to investigate whether SLD and its subclassification could stratify hepatocellular carcinoma (HCC) risk.

A cohort of 85,119 adults without viral hepatitis or heavy alcohol intake was analyzed for the risk of HCC according to SLD and its subclassification. The fibrosis-4 (FIB-4) index was used to estimate the degree of liver fibrosis.

During a median follow-up of 11.9 years, HCC was diagnosed in 123 individuals. The incidence rate of HCC per 1,000 person-years was higher in individuals with SLD than in those without SLD (0.197 vs. 0.071, p < 0.001), with an adjusted hazard ratio of 2.02 (95% confidence interval: 1.40-2.92). The HCC incidence rate per 1,000 person-years was 0, 0.180, and 0.648 for cryptogenic SLD, MASLD, and MetALD, respectively. When participants with SLD was further stratified by the FIB-4 index, the HCC incidence rate per 1,000 person-years was 0.074 for SLD with FIB-4 < 1.3 and 0.673 for SLD with FIB-4 ≥ 1.3. Of note, HCC risk was substantially high (HCC incidence rate: 1.847 per 1,000 person-years) for MetALD with FIB-4 ≥ 1.3.

HCC risk was different by SLD and its subclassification. The utilization of SLD and its subclassification can aid in stratifying HCC risk and facilitate the identification of individuals requiring interventions to mitigate the risk of HCC.

This study aimed to investigate whether metabolic dysfunction-associated fatty liver disease (MAFLD) defined by the fatty liver index (FLI) affects the decline in kidney function and whether this relationship is still observed in MAFLD defined by ultrasonography (USG). A retrospective cohort study was conducted using de-identified data from participants who received health checkups at Samsung Changwon Hospital between 2002 and 2018. The primary and secondary exposures were the presence of FLI- and USG-defined MAFLD, respectively. The primary outcome was 5-years slope of eGFR. The secondary outcome was a rapid decline in kidney function, defined as a 5-years slope of estimated glomerular filtration rate (eGFR) of less than - 3 mL/min/1.73 m2 per year. A total of 37,500 participants were included in the analysis. Participants with FLI-defined MAFLD had a larger decline in 5-year eGFR slope than those without FLI-defined MAFLD (beta coefficients - 0.11; 95% CI - 0.14 to - 0.08). Participants with FLI-defined MAFLD had a higher risk of rapid kidney function decline than those without FLI-defined MAFLD (odds ratio 1.33; 95% confidence intervals (CIs) 1.05-1.69). However, USG-defined MAFLD was less related to kidney function decline. In conclusion, the presence of FLI-defined MAFLD was associated with larger and faster kidney function decline.

Nonalcoholic fatty liver disease (NAFLD) is a common dietary disorder caused by fatty changes in the liver parenchyma and hepatocytes without alcohol consumption. The present study aimed to investigate the prevalence, characteristics, and risk factors of NAFLD in the Mashhad Persian Cohort Study population.

The present population-based cross-sectional study included all PERSIAN Organizational Cohort study in Mashhad University of Medical Sciences (POCM), Mashhad, Iran by census sampling method. Eligible participants were divided into two groups due to their NAFLD condition (NAFLD positive or NAFLD negative). All enrolled participants were evaluated based on their clinical aspects, anthropometric measures, laboratory tests, and ultrasound features. Statistical analysis was conducted using SPSS software version 16 (SPSS Inc., Chicago, USA -version 16). A P-value less than 0.05 was considered as the significance level.

A total of 1198 individuals were included in the study, of which 638 (53.3%) were male and the rest were female. The mean age of the participants was 46.89 ± 8.98 years. A total of 246 patients (20.53%) were NAFLD positive, of which 122 (49.59%) were in grade 1, 112 (45.52%) were in grade 2, and 12 (4.87%) were in grade 3. The prevalence of fatty liver was significantly higher in males than in females (p < 0.001). There were significant differences between NAFLD positive and NAFLD negative participants in terms of having a history of hypertension (P = 0.044), body mass index (P < 0.001), body fat percentage (P = 0.001), waist circumference (P < 0.001), liver craniocaudal length (P = 0.012), fasting blood sugar (FBS) (P = 0.047), aspartate aminotransferase (AST) (P = 0.007), and alanine aminotransferase (ALT) (P = 0.001). Further analysis revealed a strong significant association between BMI, previous history of hypertension, higher levels of serum ALT, and NAFLD (P < 0.05).

It can be concluded that ultrasound findings accompanied by laboratory AST and ALT level enzymes could be a cost-benefit approach for NAFLD early diagnosis. The craniocaudal size of the liver could be a beneficent marker for estimating the severity of the disease; however, more studies are recommended to evaluate this variable for future practice against the issue.

Plant-based diet (PDI) as resource of antioxidants and anti-inflammatory phytochemicals, that was considered to protect against onset and development of Non-alcoholic fatty liver disease (NAFLD).

To investigate the association between plant-based diet and NAFLD in adults.

The present case control study was conducted on 240 individuals (120 with NAFLD and 120 control) aged 20-69 years. Provided recommendations by the American College of Gastroenterology and the American Gastroenterological Association were used for NAFLD diagnosis. Dietary intake was assessed using 178-food item food frequency questionnaire (FFQ). Also, plant-based diet score was evaluated based on 18 food groups classified into animal foods, healthy and unhealthy plant foods. A multiple logistic regression model was used to examine the relationship between fatty liver disease and tertiles of PDI.

The results of this study showed that we did not observe any association between tertiles of PDI and NAFLD in crude model (OR: 1.29, 95%CI:0.66-2.52, P:0.44) and after adjustment for confounders including age, energy intake, physical activity, body mass index (OR:0.76, 95%CI: 0.31-1.86, P:0.52). Also, there were not any association of tertiles of healthy PDI (hPDI) (OR:1.14, 95%CI: 0.50-2.60, P:0.74) and unhealthy PDI (uhPDI) (OR:0.89, 95%CI:0.36-2.18, P: 0.79) with NAFLD after full adjustment for potential confounders.

There was not any association of PDI, hPDI, and uPDI with NAFLD in adults. More research needs to examine whether this specific diet may impact and improve NAFLD.

In animal models of obesity, adipocyte-derived versican, and macrophage-derived biglycan play a crucial role in mediating adipose tissue inflammation. The aim was to investigate levels of versican and biglycan in obese children and any potential association with body adipose tissue and hepatosteatosis.

Serum levels of versican, biglycan, interleukin-6 (IL-6), and high sensitivity C-reactive protein (hsCRP) were measured by ELISA. Fat deposition in the liver, spleen, and subcutaneous adipose tissue was calculated using the IDEAL-IQ sequences in magnetic resonance images. Bioimpedance analysis was performed using the Tanita BC 418 MA device.

The study included 36 obese and 30 healthy children. The age of obese children was 13.6 (7.5-17.9) years, while the age of normal weight children was 13.0 (7.2-17.9) years (p=0.693). Serum levels of versican, hsCRP, and IL-6 were higher in the obese group (p=0.044, p=0.039, p=0.024, respectively), while no significant difference was found in biglycan levels between the groups. There was a positive correlation between versican, biglycan, hsCRP, and IL-6 (r=0.381 p=0.002, r=0.281 p=0.036, rho=0.426 p=0.001, r=0.424 p=0.001, rho=0.305 p=0.017, rho=0.748 p<0.001, respectively). Magnetic resonance imaging revealed higher segmental and global hepatic steatosis in obese children. There was no relationship between hepatic fat content and versican, biglycan, IL-6, and hsCRP. Versican, biglycan, hsCRP, and IL-6 were not predictive of hepatosteatosis. Body fat percentage >32% provided a predictive sensitivity of 81.8% and a specificity of 70.5% for hepatosteatosis [area under the curve (AUC): 0.819, p<0.001]. Similarly, a body mass index standard deviation score >1.75 yielded a predictive sensitivity of 81.8% and a specificity of 69.8% for predicting hepatosteatosis (AUC: 0.789, p<0.001).

Obese children have higher levels of versican, hsCRP, and IL-6, and more fatty liver than their healthy peers.

The incidence of steatotic liver disease has increased in recent years. Thus, steatotic liver disease is a major public health issue in Japan. This study investigated the association between weight reduction and the remission of metabolic dysfunction-associated steatotic liver disease (MASLD)/Metabolic and alcohol related/associated liver disease (MetALD) in Japanese individuals undergoing health checkups.

This retrospective observational study included 8,707 Japanese patients with MASLD/MetALD who underwent health checkups from May 2015 to March 2023. The participants were monitored for its remission at their subsequent visit. MASLD was diagnosed on abdominal ultrasonography and based on the presence of at least one of five metabolic abnormalities. The impact of body mass index (BMI) reduction on MASLD/MetALD remission was assessed via logistic regression analysis and using receiver operating characteristic curves.

Logistic regression analysis revealed that weight loss was significantly associated with MASLD/MetALD remission. Other factors including exercise habits and reduced alcohol consumption were significant predictors of MASLD/MetALD remission in the overall cohort and in male patients. The optimal BMI reduction cutoff values for MASLD/MetALD remission were 0.9 kg/m2 and 4.0% decrease in the overall cohort, 0.85 kg/m2 and 3.9% decrease in males, and 1.2 kg/m2 and 4.5% decrease in females. In participants with a BMI of 23 kg/m2, the cutoff values were 0.75 kg/m2 and 2.7% BMI reduction.

Weight reduction plays an important role in both MASLD and MetALD remission among Japanese individuals. That is, targeting specific BMI reduction is effective. This underscores the importance of targeted weight management strategies in preventing and managing MASLD/MetALD in the Japanese population.

Metabolic dysfunction-associated fatty liver disease (MAFLD) is a new term proposed to replace non-alcoholic fatty liver disease (NAFLD). We analyzed the ultrasonographic findings of MAFLD and NAFLD.

We conducted a retrospective cross-sectional study of subjects aged ≥19 years who underwent a health screening examination, including ultrasonography, (n = 17,066). Patients were separated into one of three groups; pure MAFLD (n = 5304), pure NAFLD (n = 579), and both NAFLD & MAFLD (n = 11,183). The outcomes were the degree of fatty liver disease and liver cirrhosis, defined by ultrasonography. In addition, the risk of ultrasonographic cirrhosis was assessed in the MAFLD group based on clinical characteristics.

The pure NAFLD group had a lower risk of severe fatty liver disease than the both NAFLD & MAFLD groups (0.9 % vs. 4.4 %, p < 0.001). Cirrhosis was not diagnosed in the NAFLD group. Cirrhosis was more common in the pure MAFLD group than in the both NAFLD & MAFLD group (0.3 % vs. 0.0 %, p < 0.001). In the MAFLD group, multivariable analysis showed that diagnosis by hepatic steatosis index (Odds ratio [OR], 12.39; 95 % confidence interval [CI], 3.40-45.19; p < 0.001) or significant alcohol intake (OR, 9.58, 95 % CI, 1.93-47.61; p = 0.006) was independently associated with risk of liver cirrhosis on ultrasonography.

Liver cirrhosis was more frequently identified on ultrasonography in patients with MAFLD than in NAFLD. MAFLD diagnosed using the hepatic steatosis index or significant alcohol intake is a risk factor for liver cirrhosis.

Depressive-like behavior has been shown to be associated with liver damage. This study aimed to evaluate the impact of three different models of depression on the behavior of mice with liver injury.

During the 4 weeks of methionine/choline deficiency diet (MCD), adult C57BL/6 mice were randomly divided into four groups: MCD (no stress protocol, n = 6), chronic unpredictable mild stress (CUMS, n = 9), acute and repeated forced swim stress [aFSS (n = 9) and rFSS (n = 9)].

All depression protocols induced increased anhedonia and anxiety-like behavior compared to baseline and had no impact on the severity of liver damage, according to ultrasonography. However, different protocols evoked different overall behavior patterns. After the depressive-like behavior induction protocols, animals subjected to aFSS did not exhibit anxiety-like behavior differences compared to MCD animals, while mice subjected to CUMS showed additional weight loss compared to FSS animals. All tested protocols for inducing depressive-like behavior decreased the short-term memory of mice with liver damage, as assessed by the novel object recognition test (NORT).

Our results show that the use of all protocols seems to generate different levels of anxiety-like behavior, but only the depressive-like behavior induction procedures associate additional anhedonia and memory impairment in mice with liver injury.

To explore the feasibility of a 5G-based telerobotic ultrasound (US) system for providing qualified abdominal US services on a rural island.

This prospective study involved two medical centers (the tele-radiologist site's hospital and the patient site's hospital) separated by 72 km. Patients underwent 5G-based telerobotic US by tele-radiologists and conventional US by on-site radiologists from September 2020 to March 2021. The clinical feasibility and diagnostic performance of the 5G-based telerobotic abdominal US examination were assessed based on safety, duration, image quality, diagnostic findings, and questionnaires.

A total of 401 patients (217 women and 184 men; mean age, 54.96 ± 15.43 years) were enrolled. A total of 90.1% of patients indicated no discomfort with the telerobotic US examination. For the examination duration, telerobotic US took longer than conventional US (12.54 ± 3.20 min vs. 7.23 ± 2.10 min, p = 0.001). For image quality scores, the results of the two methods were similar (4.54 ± 0.63 vs. 4.57 ± 0.61, p = 0.112). No significant differences were found between the two methods in measurements for the aorta, portal vein, gallbladder, kidney (longitudinal diameter), prostate, and uterus; however, telerobotic US underestimated the transverse diameter of the kidney (p < 0.05). A total of 504 positive results, including 31 different diseases, were detected. Among them, 455 cases were identified by the two methods; 17 cases were identified by telerobotic US only; and 32 cases were identified by conventional US only. There was good consistency in the diagnosis of 29 types of disease between the two methods (κ = 0.773-1.000). Furthermore, more than 90% of patients accepted the telerobotic US examination and agreed to pay additional fees in future.

The 5G-based telerobotic US system can expand access to abdominal US services for patients in rural areas, thereby reducing health care disparities.

Non-alcoholic fatty liver disease (NAFLD) is characterized by accumulation of fat in the liver cells caused by means other than alcohol consumption. It is one of the most common chronic liver diseases worldwide and can lead to severe conditions, such as cirrhosis and liver cancer. NAFLD is often associated with other metabolic disorders, such as obesity and diabetes, and is closely related to lifestyle factors, such as diet and physical activity [1]. The diagnosis as well as management of NAFLD are complex and involve a multidisciplinary approach. The available treatment options include lifestyle modifications, pharmacological interventions, and in severe cases, liver transplantation. The increasing prevalence of NAFLD highlights the urgent requirement of effective prevention and management strategies. This disease is a growing health concern in India, given the rise in the incidence of obesity and diabetes. According to a study published in the Journal of Clinical and Experimental Hepatology in 2020, the prevalence of NAFLD in India is estimated to be between 9% and 32%. In accordance with the research population and diagnostic criteria employed, a study published in the Indian Journal of Gastroenterology in 2019 found that the prevalence of NAFLD in India ranged from 9.6% to 32.3% [2]. The same study also revealed that the prevalence of non-alcoholic steatohepatitis (NASH), a highly severe form of NAFLD, ranged from 1.5% to 8.4%. These statistics highlight the need for increased awareness and preventive measures to manage the growing burden of NAFLD in India.

Integrating deep learning with clinical expertise holds great potential for addressing healthcare challenges and empowering medical professionals with improved diagnostic tools. However, the need for annotated medical images is often an obstacle to leveraging the full power of machine learning models. Our research demonstrates that by combining synthetic images, generated using diffusion models, with real images, we can enhance nonalcoholic fatty liver disease (NAFLD) classification performance even in low-data regime settings. We evaluate the quality of the synthetic images by comparing two metrics: Inception Score (IS) and Fréchet Inception Distance (FID), computed on diffusion- and generative adversarial network (GAN)-generated images. Our results show superior performance for the diffusion-generated images, with a maximum IS score of 1.90 compared to 1.67 for GANs, and a minimum FID score of 69.45 compared to 100.05 for GANs. Utilizing a partially frozen CNN backbone (EfficientNet v1), our synthetic augmentation method achieves a maximum image-level ROC AUC of 0.904 on a NAFLD prediction task.

Several body components are known to be associated with non-alcoholic fatty liver disease (NAFLD) in children. However, the relative contributions of soft tissue mass components as risk or protective factors of NAFLD are largely unknown because measurements of these components are often highly correlated. Therefore, we aimed to estimate levels of association between soft tissue mass components and NAFLD.

We collected the medical records of 555 Chinese children (aged 3-18 years). Five mutually exclusive and exhaustive components of soft tissue mass were measured using dual energy X-ray absorptiometry. NAFLD was diagnosed with abdominal B-ultrasound scan. We fit Dirichlet regression and multivariate linear regression models wherein age and NAFLD were used as predictors of the proportional measurements of soft tissue mass components.

The proportion of android fat was significantly higher in children with NAFLD than in those without NAFLD (ratio of proportions ranged from 1.18 to 1.30), whereas proportions of trunk lean and limb lean were significantly lower (ratio of proportions ranged from 0.87 to 0.92 for trunk lean and from 0.82 to 0.91 for limb lean). The proportion of gynoid fat was slightly higher in boys with NAFLD than in those without NAFLD (ratio = 1.05), but this proportion was not significantly higher in girls. The association between the proportion of android fat and NAFLD appeared to be somewhat greater than the associations between proportions of trunk lean or limb lean components and NAFLD.

Our findings suggest that lowering fat mass and increasing lean mass can both be used to combat NAFLD in children and that more studies are needed to determine the association between gynoid fat and NAFLD.

Scant West African data on non-alcoholic fatty liver disease (NAFLD) means there is little representation of this population in the modelling used to derive biomarkers and predictive indices for risk stratification of patients for the presence of hepatic steatosis. This study evaluates the performance of the fatty liver index (FLI), hepatic steatosis index (HSI) and triglyceride-glucose (TyG) index and its derivatives in predicting ultrasound detected NAFLD in a locally resident population of Ghanaian participants.

A post hoc analysis of data from a cross sectional assessment of NAFLD and cardiovascular risk was performed. Data from 210 participants without significant alcohol intake, or secondary causes of fatty liver and not on steatogenic drugs was evaluated. A structured questionnaire had been used to collect demographic data, medical and drug history. Anthropometry, blood sampling for liver chemistry and fasting lipids were performed. Hepatic steatosis was detected by ultrasonography. A retrospective analysis involving multivariate binary logistic regression assessed FLI, HIS, TyG (and its derivatives) as predictors of NAFLD with p < .05 considered statistically significant. Sensitivity, specificity, predictive values, likelihood ratios were calculated and accuracy of the proxies evaluated from area under the receiver operating characteristics curve (AUROC). All the biomarkers and indices were significantly associated with NAFLD (p ≤ .001). All the lipid and fatty liver indices assessed performed acceptably as predictors of NAFLD. FLI (AUC = 0.8, 95% CI [0.74-0.87]), TyG-WC (AUC = 0.81, 95% CI [0.75-0.88]) and TyG-WHtR (AUC = 0.81, 95% CI [0.74-0.88]) performed best at predicting NAFLD. Whilst in all cases the markers had good specificity (>90%) they lacked sufficient sensitivity with FLI having the highest sensitivity of 36.7%. Their overall accuracy was greater than 70% in each case.

The overall accuracy of HSI, FLI, TyG index and its derivatives (TyG WHtR, TyG BMI, TyG WC) was acceptable for predicting NAFLD in this population. Given their performance in this study and in light of their low cost, accessibility, easy interpretation and non-invasive nature; they are suitable tools for screening in the Ghanaian population.

Adipose insulin resistance (Adipo-IR) is associated with multiple metabolic diseases, including non-alcoholic fatty liver disease (NAFLD). The study aimed to evaluate sex differences in the association between Adipo-IR and NAFLD, and further investigated other potential modifiers.

This cross-sectional study enrolled adults without diabetes who underwent physical examinations in Beijing Chao-Yang Hospital. We calculated the Adipo-IR index as the product of the fasting insulin and free fatty acid concentration. We categorized Adipo-IR into four groups according to quartiles, using the first interquartile range (Q1) as the reference. Logistic regression was used stratified by the modifiers after adjustment for potential confounders.

There were 5586 participants in the study, 49.8% (n = 2781) of whom were women and 30.4% (n = 1698) with NAFLD. There was a graded positive association between Adipo-IR and NAFLD, with sex (P = 0.01) and hyperlipidemia (P = 0.02) modifying this association. In the hyperlipidemic women, for one unit increase in log-Adipo-IR, the odds of having NAFLD increased by 385% after adjustment for potential confounders (OR = 4.85, 95%CI 3.54-6.73, P < 0.001). However, it turned out that the odds of having NAFLD increased by 131% (OR = 2.31, 95%CI 1.74-3.11, P < 0.001), 216% (OR = 3.16, 95%CI 2.56-3.93, P < 0.001), 181% (OR = 2.81, 95%CI 1.88-4.28, P < 0.001) in normolipidemic men, hyperlipidemic men, and normolipidemic women, respectively. Similarly, the ORs for the association between Adipo-IR and NAFLD in women with age ≥ 50 years were higher than ORs in women with age < 50 years.

The positive correlation between Adipo-IR and NAFLD was stronger in hyperlipidemic women, compared with normolipidemic or hyperlipidemic men, or normolipidemic women. The association also strengthened for women over 50 years. Treatment strategies targeting Adipo-IR to alleviate NAFLD may be of value, especially in hyperlipidemic women after menopause.

Point-of-care ultrasound (POCUS) refers to the use of ultrasound imaging through pocket-sized sonographic devices at the patient's bedside, to make a diagnosis or direct a procedure and immediately answer a clinical question. Its goal is to broaden the physical examination, not to replace conventional ultrasound studies. POCUS has evolved as a complement to physical examination and has been adopted by different medical specialties, including hepatology. A narrative synthesis of the evidence on the applications of POCUS in hepatology was carried out, describing its usefulness in the diagnosis of cirrhosis of the liver, metabolic dysfunction-associated steatotic liver disease (MASLD), decompensated cirrhosis, and portal hypertension. The review also encompasses more recent applications in the hemodynamic evaluation of the critically ill patient with cirrhosis of the liver, patients with other liver diseases, as well as in the ultrasound guidance of procedures. POCUS could make up part of the daily clinical practice of gastroenterologists and hepatologists, simplifying the initial evaluation of patients and optimizing clinical management. Its accessibility, ease of use, and low adverse event profile make POCUS a useful tool for the properly trained physician in the adequate clinical setting. The aim of this review was to describe the available evidence on the usefulness of POCUS in the daily clinical practice of gastroenterologists and hepatologists.

The aims of the work described here were to assess shear wave attenuation (SWA) in volunteers and patients with non-alcoholic fatty liver disease (NAFLD) and compare its diagnostic performance with that of shear wave dispersion (SWD), magnetic resonance imaging (MRI) proton density fat fraction (PDFF) and biopsy.

Forty-nine participants (13 volunteers and 36 NAFLD patients) were enrolled. Ultrasound and MRI examinations were performed in all participants. Biopsy was also performed in patients. SWA was used to assess histopathology grades as potential confounders. The areas under curves (AUCs) of SWA, SWD and MRI-PDFF were assessed in different steatosis grades by biopsy. Youden's thresholds of SWA were obtained for steatosis grading while using biopsy or MRI-PDFF as the reference standard.

Spearman's correlations of SWA with histopathology (steatosis, inflammation, ballooning and fibrosis) were 0.89, 0.73, 0.62 and 0.31, respectively. Multiple linear regressions of SWA confirmed the correlation with steatosis grades (adjusted R2 = 0.77, p < 0.001). The AUCs of MRI-PDFF, SWA and SWD were respectively 0.97, 0.99 and 0.94 for S0 versus ≥S1 (p > 0.05); 0.94, 0.98 and 0.78 for ≤S1 versus ≥S2 (both MRI-PDFF and SWA were higher than SWD, p < 0.05); and 0.90, 0.93 and 0.68 for ≤S2 versus S3 (both SWA and MRI-PDFF were higher than SWD, p < 0.05). SWA's Youden thresholds (Np/m/Hz) (sensitivity, specificity) for S0 versus ≥S1, ≤S1 versus ≥S2 and ≤S2 versus S3 were 1.05 (1.00, 0.92), 1.37 (0.96, 0.96) and 1.51 (0.83, 0.87), respectively. These values were 1.16 (1.00, 0.81), 1.49 (0.91, 0.82) and 1.67 (0.87, 0.92) when considering MRI-PDFF as the reference standard.

In this pilot study, SWA increased with increasing steatosis grades, and its diagnostic performance was higher than that of SWD but equivalent to that of MRI-PDFF.

Objective To predict fatty liver disease (FLD), including nonalcoholic FLD (NAFLD) and metabolic dysfunction-associated FLD (MAFLD), from blood tests and anthropometric measurements, the fatty liver index (FLI) and triglyceride glucose-body mass index (TyG-BMI) have been reported as promising indicators. We evaluated the predictive ability of several indices, including the waist circumference, BMI, FLI and TyG-BMI, that might predict FLD in non-obese individuals undergoing health checkups. Methods This retrospective observational study enrolled non-obese subjects who underwent abdominal ultrasonography between May 1, 2015, and June 30, 2022. Obesity was defined as a BMI <25 kg/m2. FLD was diagnosed by abdominal ultrasonography. Using a receiver operating characteristic analysis, we examined the predictive validity of indices for NAFLD and MAFLD by calculating the area under the curve (AUC). Results Of the 24,825 subjects (mean age 44.3±10.0 years old; 54% men) enrolled in this examination of the association of indices, including FLI and TyG-BMI, with NAFLD, NAFLD was diagnosed in 3,619 (27%) men and 733 (6%) women. In both men and women, the FLI and TyG-BMI had significantly higher AUC values for NAFLD prediction than the other indicators (FLI: 0.786 for men and 0.875 for women, TyG-BMI: 0.783 for men and 0.868 for women). In analyses of subjects with a BMI <23 kg/m2, the superiority of the FLI and TyG-BMI remained unchanged. The FLI and TyG-BMI also had significantly higher AUC values for MAFLD prediction than the other indicators. Conclusion The FLI and TyG-BMI had a particularly high predictive ability for NAFLD and MAFLD in non-obese subjects.

Follicle-stimulating hormone (FSH) was negatively associated with nonalcoholic fatty liver disease (NAFLD) in women older than 55 years old. People with obesity and diabetes had higher prevalence of NAFLD. Thus, we aimed to explore the association between FSH and NAFLD in postmenopausal women with type 2 diabetes mellitus (T2DM).

A total of 583 postmenopausal women with T2DM with an average age of 60.22 ± 6.49 were recruited in this cross-sectional study through January 2017 to May 2021. Anthropological data, biochemical indexes, and abdominal ultrasound results were retrospectively collected. Abdominal ultrasound was used to diagnose NAFLD. FSH was measured by enzymatic immunochemiluminescence and divided into tertiles for further analysis. The logistic regression was used to assess the association of FSH with prevalent NAFLD. Likelihood ratio tests were used to assess the interactions between groups.

A total of 332 (56.94%) postmenopausal women had NAFLD. Compared with postmenopausal women in the lowest tertile of FSH, postmenopausal women in the highest tertile of FSH had lower prevalence of NAFLD (p < .01). After adjusting for age, diabetes duration, metabolism-related indicators, and other sex-related hormones, FSH was inversely associated with NAFLD (odds ratio: 0.411, 95% confidence intervals: 0.260-0.651, p < .001). In subgroup analysis, there were no significant interactions of FSH with strata of metabolic factors on the association of NAFLD.

FSH was negatively and independently associated with NAFLD in postmenopausal women with type 2 diabetes mellitus. It might be a potential index for screening and identifying individuals with high risk of NAFLD in postmenopausal women.

Using ultrasound imaging to diagnose liver steatosis is of great significance for preventing diseases such as cirrhosis and liver cancer. Accurate diagnosis under conditions of low quality, noise and poor resolutions is still a challenging task. Physiological studies have shown that the visual cortex of the biological visual system has selective attention neural mechanisms and feedback regulation of high features to low features. When processing visual information, these cortical regions selectively focus on more sensitive information and ignore unimportant details, which can effectively extract important features from visual information. Inspired by this, we propose a new diagnostic network for hepatic steatosis. In order to simulate the selection mechanism and feedback regulation of the visual cortex in the ventral pathway, it consists of a receptive field feature extraction module, parallel attention module and feedback connection. The receptive field feature extraction module corresponds to the inhibition of the non-classical receptive field of V1 neurons on the classical receptive field. It processes the input image to suppress the unimportant background texture. Two types of attention are adopted in the parallel attention module to process the same visual information and extract different important features for fusion, which improves the overall performance of the model. In addition, we construct a new dataset of fatty liver ultrasound images and validate the proposed model on this dataset. The experimental results show that the network has good performance in terms of sensitivity, specificity and accuracy for the diagnosis of fatty liver disease.

The prevalence of non-alcoholic fatty liver disease (NAFLD) is increasing in the USA. Some of these patients develop non-alcoholic steatohepatitis (NASH), which can progress to cirrhosis. Ultrasound imaging is one of the most used modalities for diagnosing hepatic steatosis. Primary care providers are increasingly using point-of-care ultrasound (POCUS), which could increase the number of subjects diagnosed with NAFLD. This study evaluates the accuracy of POCUS in identifying patients with NASH.

Patients with hepatic steatosis without excess alcohol intake or other liver diseases undergoing liver biopsy were included in this study. These patients underwent POCUS and vibration-controlled transient elastography (VCTE) evaluations within 3 months of a liver biopsy. A comparison of POCUS data with liver histology and VCTE were made to assess the validity of POCUS evaluation in diagnosing NAFLD and NASH.

The steatosis score from the liver histology had a low correlation with the controlled attenuation parameter score from VCTE ( r  = 0.27) and a moderate correlation with the grade of steatosis detected by the POCUS exam ( r  = 0.57). The NAFLD activity score on histology was found to correlate with the ultrasonographic fatty liver index (USFLI) from the POCUS exam ( r  = 0.59). A USFLI ≥ 6 diagnosed NASH with a sensitivity of 81%, and a value of ≤3 ruled out the diagnosis of NASH with a sensitivity of 100%.

The provider can use the POCUS exam in clinical practice to diagnose NAFLD and reliably stratify patients who have NASH.

The aim of this study was to identify characteristics of non-alcoholic fatty liver disease (NAFLD) in adults with HFE p.C282Y/p.C282Y.

We retrospectively studied non-Hispanic white hemochromatosis probands with iron overload (serum ferritin (SF) > 300 µg/L (M), > 200 µg/L (F)) and p.C282Y/p.C282Y at non-screening diagnosis who did not report alcohol consumption > 14 g/d, have cirrhosis or other non-NAFLD liver disorders, use steatogenic medication, or have diagnoses of heritable disorders that increase NAFLD risk. We identified NAFLD-associated characteristics using univariate and multivariable analyses.

There were 66 probands (31 men, 35 women), mean age 49 ± 14 (SD) y, of whom 16 (24.2%) had NAFLD. The following characteristics were higher in probands with NAFLD: median SF (1118 µg/L (range 259, 2663) vs. 567 µg/L (247, 2385); p = 0.0192); prevalence of elevated ALT/AST (alanine/aspartate aminotransferase) (43.8% vs. 10.0%; p = 0.0056); and prevalence of type 2 diabetes (T2DM) (31.3% vs. 10.0%; p = 0.0427). Mean age, sex, and prevalences of human leukocyte antigen-A*03 positivity, body mass index ≥ 30.0 kg/m², hyperlipidemia, hypertension, and metabolic syndrome in probands with/without NAFLD did not differ significantly. Logistic regression on NAFLD using variables SF, elevated ALT/AST, and T2DM revealed: SF (p = 0.0318; odds ratio 1.0-1.0) and T2DM (p = 0.0342; 1.1-22.3). Median iron removed to achieve iron depletion (QFe) in probands with/without NAFLD did not differ significantly (3.6 g (1.4-7.2 g) vs. 2.8 g (0.7-11.0 g), respectively; p = 0.6862).

NAFLD in hemochromatosis probands with p.C282Y/p.C282Y is associated with higher median SF and greater T2DM prevalence, after adjustment for other factors. NAFLD does not influence QFe significantly.

Non-alcoholic fatty liver disease (NAFLD) is the most common chronic liver disease, which can progress from simple steatosis to advanced cirrhosis and hepatocellular carcinoma. Clinical diagnosis of NAFLD is crucial in the early stages of the disease. The main aim of this study was to apply machine learning (ML) methods to identify significant classifiers of NAFLD using body composition and anthropometric variables. A cross-sectional study was carried out among 513 individuals aged 13 years old or above in Iran. Anthropometric and body composition measurements were performed manually using body composition analyzer InBody 270. Hepatic steatosis and fibrosis were determined using a Fibroscan. ML methods including k-Nearest Neighbor (kNN), Support Vector Machine (SVM), Radial Basis Function (RBF) SVM, Gaussian Process (GP), Random Forest (RF), Neural Network (NN), Adaboost and Naïve Bayes were examined for model performance and to identify anthropometric and body composition predictors of fatty liver disease. RF generated the most accurate model for fatty liver (presence of any stage), steatosis stages and fibrosis stages with 82%, 52% and 57% accuracy, respectively. Abdomen circumference, waist circumference, chest circumference, trunk fat and body mass index were among the most important variables contributing to fatty liver disease. ML-based prediction of NAFLD using anthropometric and body composition data can assist clinicians in decision making. ML-based systems provide opportunities for NAFLD screening and early diagnosis, especially in population-level and remote areas.

Nonalcoholic fatty liver disease (NAFLD) is one of major causes of end-stage liver disease in the coming decades, but it shows few symptoms until it develops into cirrhosis. We aim to develop classification models with machine learning to screen NAFLD patients among general adults. This study included 14,439 adults who took health examination. We developed classification models to classify subjects with or without NAFLD using decision tree, random forest (RF), extreme gradient boosting (XGBoost) and support vector machine (SVM). The classifier with SVM was showed the best performance with the highest accuracy (0.801), positive predictive value (PPV) (0.795), F1 score (0.795), Kappa score (0.508) and area under the precision-recall curve (AUPRC) (0.712), and the second top of area under receiver operating characteristic curve (AUROC) (0.850). The second-best classifier was RF model, which was showed the highest AUROC (0.852) and the second top of accuracy (0.789), PPV (0.782), F1 score (0.782), Kappa score (0.478) and AUPRC (0.708). In conclusion, the classifier with SVM is the best one to screen NAFLD in general population based on the results from physical examination and blood testing, followed by the classifier with RF. Those classifiers have a potential to screen NAFLD in general population for physician and primary care doctors, which could benefit to NAFLD patients from early diagnosis.