【】Objective:Based on multimodal ultrasound and clinical data

a prediction model of cervical lymph node metastasis (CLNM) in patients with papillary thyroid microcarcinoma (PTMC) was established.Methods:Eighty-two PTMC patients diagnosed and treated in our hospital from January

2022 to March

2023 were examined by conventional ultrasound

contrast-enhanced ultrasound and microvascular imaging (SMI) before operation

and were divided into CLNM group (32 cases) and non-CLNM group (50 cases) according to the presence or absence of CLNM. The multimodal ultrasound data and clinical data of the two groups were compared

and the risk factors of CLNM in PTMC patients were analyzed. The CLNM prediction of PTMC patients was constructed based on multimodal ultrasound and clinical characteristics.Result:Compared with the non-CLNM group

CLNM group had significant differences in age

maximum diameter of focus

number of microcalcifications

enhancement of early capsule continuity and thyroid infiltration (P<0.05). According to binary Logistic regression analysis

older age is the protective factor of CLNM in PTMC patients (P<0.05)

and the largest diameter of lesions > 1.0cm

the number of microcalcifications > 5

the interruption of early capsule enhancement and thyroid infiltration are the risk factors of CLNM in PTMC patients (P<0.05). Hosmer-Lemeshow test shows that χ2=6.868

P=0.551

and the model has a good goodness of fit. ROC analysis showed that the AUC

Jordan index

sensitivity and specificity of the prediction model of CLNM in PTMC patients based on multimodal ultrasound and clinical features were 0.924

0.733

81.3%

92.0%

95%CI (0.864

0.984) respectively

and the accuracy was 87.8%.Conclusion:The prediction model of CLNM in PTMC patients based on multimodal ultrasound and clinical features in this study has good prediction value and can provide auxiliary reference for the prediction of CLNM in PTMC patients.