Evaluation of the diagnostic value of CEUS LI-RADS combined with artificial intelligence assisted decision model for hepatocellular carcinoma
Objective To explore the diagnostic value of contrast-enhanced ultrasound liver imaging reporting and data system(CEUS LI-RADS)combined with artificial intelligence for hepatocellular carcinoma.Methods A total of 88 patients with hepatocellular carcinoma were selected and underwent contrast-enhanced ultrasound(CEUS)examination after admission.CEUS LI-RADS was used to evaluate the malignancy of hepatocellular carcinoma and artificial intelligence was used for diagnosis based on CEUS results.The individual and joint diagnostic results of CEUS LI-RADS and artificial intelligence for hepatocellular carcinoma were compared,and the diagnostic value of CEUS LI-RADS for different pathological characteristics of hepatocellular carcinoma were analyzed.Results The combined diagnosis of CEUS LI-RADS and artificial intelligence has higher sensitivity and higher accuracy for the diagnosis of hepatocellular carcinoma than the single protocol diagnosis,with lower missed diagnosis rate,and differerce was statistically significant(P<0.05).There were significant differences in CEUS LI-RADS classification among LR-1～LR-3,LR-4 and LR-5 in patients with liver cell tumors<5 cm positive(hepatocellular carcinoma)and negative(benign stem cell tumor),and there was a significant difference between LR-1～LR-3 and LR-5 in patients with liver cell tumors≥5 cm positive and negative(P<0.05);there were significant differences in CEUS LI-RADS classification among LR-3,LR-4 and LR-5 in patients with hepatocellular carcinoma of different degrees of differentiation(P<0.05).Conclusion The combination of CEUS LI-RADS and artificial intelligence assisted decision model can be used in the diagnosis of hepatocellular carcinoma with higher accuracy and sensitivity,and can improve the diagnostic efficiency and lower the rate of missed diagnosis.
contrast-enhanced ultrasound liver imaging reporting and data systemartificial intelligence assisted decision-makinghepatocellular carcinomadiagnostic value
NETL
NSTL
万方数据
