To analyze the liver sonographic features of middle-aged and elderly patients with nonalcoholic fatty liver disease (NAFLD), and explore the correlation between these sonographic features and blood lipids. Methods 120 middle-aged and elderly patients with NAFLD admitted to the hospital from July 2019 to July 2021 were selected as the research subjects. They were divided into control group (n=28, normal blood lipids) and observation group (n=92, dyslipidemia) according to the dyslipidemia guidelines. Liver sonographic features of the two groups were compared, and a diagnostic model was established based on these features after logistic regression analysis. Pearson correlation coefficient was used to analyze the relationship between sonographic features and blood lipids. Results There were statistically significant differences in liver size and shape, distant echo attenuation, intrahepatic duct structure and color blood flow signal between the two groups (P＜0.05). The model equation obtained based on logistic regression analysis was as follows: Log(p)=-0.938+1.725*A1-1.549*A2+1.263*A3+1.371*A4 (A1=liver enlargement, blunt liver surface angle, A2=echo attenuation, A3= thin and unclear hepatic veins, and A4=reduced or not obvious color flow signal). The AUC of this model to predict dyslipidemia in middle-aged and elderly patients with NAFLD was 0.925 (95% CI: 0.862-0.965). When the Youden index was 0.723 and the critical value was 0.281, the sensitivity and specificity of this model were 90.22% and 82.14%. The ultrasound scores were positively correlated with triglycerides (TG), total cholesterol(TC), and low-density lipoprotein cholesterol(LDL-C) (P＜0.05), and negatively correlated with high-density lipoprotein cholesterol(HDL-C) (P＜0.05). Conclusion Blood lipids in middle-aged and elderly patients with NAFLD are closely related to liver sonographic features. A quantitative standard of dyslipidemia in middle-aged and elderly patients with NAFLD can be obtained by establishing a diagnostic model, thereby improving the diagnostic accuracy, and providing a basis for subsequent diagnosis and treatment