Objective To explore the differential diagnosis and clinical significance of B-mode dual-mode ultrasound imaging in patients with hyperuricemia complicated with gouty arthritis and provide help for clinical diagnosis and treatment. Methods 41 patients with HUA complicated with GA in our hospital from January 2018 to December 2022 were included in the GA group. Eighteen non-GA patients with joint swelling and pain in our hospital in the same period were included in the non-GA group. The general clinical data and laboratory indexes of the patients were collected and compared. The elastic and B-mode ultrasound images of the patients were collected, and the bimodal imaging quantitative features were extracted. Each mode includes morphological features, image intensity features and GLCM features. The LASSO regression 50% discount cross-validation method were used to screen the imaging features and clinical features included in B mode, real-time elastography (RTE) mode and all their features, respectively. The corresponding features when tuning the parameter lambda.1se are selected, and the optimal feature subset of imaging is obtained. Then support vector machine is used to diagnose GA and non-GA classification on the imaging feature subset of each modal. Finally, Adaboost algorithm was used to fuse and evaluate the classification results of different modes and different feature subsets. Results 14 non-zero coefficient features were screened based on B-mode ultrasound images, 18 non-zero coefficient features were screened based on RTE ultrasound images, and 16 non-zero coefficient features were selected by series fusion of B-type features and RTE features. The best ACC and F1 values of type B optimal feature classification were 70.29%±3.25% and 71.30% ±4.46%, respectively, and the best ACC and F1 values of RTE feature classification were 68.01%±4.31% and 66.48%±7.23%, respectively. The best ACC and F1 values of series fusion feature classification were 74.17%±3.72% and 74.56%±5.22%, respectively. The ACC and F1 values of series fusion features are at least 3.88% and 3.26% higher than those of individual B-type features, and at least 6.61% and 8.08% higher than those of RTE features. After the Adaboost algorithm fuses the SVM classifier, the best ACC and F1 values of the whole group are 76.24%±2.50% and 75.73%±3.22%, respectively. Compared with the SVM primary classification model, the ACC and F1 values of the whole group are increased by at least 2.07% and 1.17%, respectively. ROC curve analysis showed that the AUC of class 0 and class 1 as positive samples were 0.718 and 0.910 respectively. Conclusion Based on the imaging characteristics of B-mode and RTE ultrasound, HUA combined with GA can be quantitatively characterized and effectively predicted, which has potential in the early diagnosis of GA. The classification effect of single B mode is better than that of single RTE mode, and the classification effect of dual mode is better than that of single mode.