Objective:To investigate the diagnostic efficacy of artificial intelligence(AI)software combined with contrast-enhanced ultrasound(CEUS)on portal hypertension of cirrhosis.Methods:A retrospective analysis was conducted on 106 patients with cirrhosis who admitted to Handan Communicable Disease Hospital from January 2024 to June 2024 and were confirmed by surgical pathology.In them,60 cases with positive portal hypertension were included in positive group,and 46 cases with negative portal hypertension were included in negative group.The receiver operating characteristic(ROC)curve was used to analyze the diagnostic values of ultrasound characteristics via preoperative AI quantization,enhanced ultrasound mode and the combination of them for portal hypertension of cirrhosis.Results:The mean peak intensity,mean vessel density and branching morphology index of AI software were respectively 201.23±10.18,0.62±0.11 and 1.23±0.24 in diagnosing portal hypertension of positive group,all of which were significantly higher than those of negative group(t=8.551,3.707,2.054,P<0.05).However,the mean peak reaching time,uniformity of flow perfusion,mean flow velocity and uniformity of microcirculation perfusion of the positive group were respectively(18.22±2.18)seconds,(75.22±5.46)%,(10.09±2.22)cm/s,(80.78±5.11)%,all of which were significantly lower than those in the negative group(t=3.729,5.237,4.969,4.491,P<0.05).The enhancement speed and degree of liver parenchyma,the enhancement speed of portal vein,and the enhancement speed and degree of spleen in CEUS of positive group were significantly higher than those of negative group,and the differences were significant(t=4.770,12.781,7.313,5.501,4.645,7.513,P<0.05).ROC curve analysis showed that the area under curve(AUC)value of CEUS combined with AI algorithm was 0.926 in diagnosis,which was significantly better than single CEUS and single AI analysis(0.853,0.925).Conclusion:CEUS technique combines with AI software can show higher diagnostic efficacy in the diagnosis for portal hypertension of cirrhosis,which diagnostic effect was better than single either of two methods.

Objective:To investigate the diagnostic efficacy of artificial intelligence(AI)software combined with contrast-enhanced ultrasound(CEUS)on portal hypertension of cirrhosis.Methods:A retrospective analysis was conducted on 106 patients with cirrhosis who admitted to Handan Communicable Disease Hospital from January 2024 to June 2024 and were confirmed by surgical pathology.In them,60 cases with positive portal hypertension were included in positive group,and 46 cases with negative portal hypertension were included in negative group.The receiver operating characteristic(ROC)curve was used to analyze the diagnostic values of ultrasound characteristics via preoperative AI quantization,enhanced ultrasound mode and the combination of them for portal hypertension of cirrhosis.Results:The mean peak intensity,mean vessel density and branching morphology index of AI software were respectively 201.23±10.18,0.62±0.11 and 1.23±0.24 in diagnosing portal hypertension of positive group,all of which were significantly higher than those of negative group(t=8.551,3.707,2.054,P<0.05).However,the mean peak reaching time,uniformity of flow perfusion,mean flow velocity and uniformity of microcirculation perfusion of the positive group were respectively(18.22±2.18)seconds,(75.22±5.46)%,(10.09±2.22)cm/s,(80.78±5.11)%,all of which were significantly lower than those in the negative group(t=3.729,5.237,4.969,4.491,P<0.05).The enhancement speed and degree of liver parenchyma,the enhancement speed of portal vein,and the enhancement speed and degree of spleen in CEUS of positive group were significantly higher than those of negative group,and the differences were significant(t=4.770,12.781,7.313,5.501,4.645,7.513,P<0.05).ROC curve analysis showed that the area under curve(AUC)value of CEUS combined with AI algorithm was 0.926 in diagnosis,which was significantly better than single CEUS and single AI analysis(0.853,0.925).Conclusion:CEUS technique combines with AI software can show higher diagnostic efficacy in the diagnosis for portal hypertension of cirrhosis,which diagnostic effect was better than single either of two methods.