基于灰阶超声构建症状性颈动脉斑块影像组学模型的临床研究OACSTPCD

Objective:To explore the feasibility and application value of constructing symptomatic carotid plaque radiomics model based on gray-scale ultrasound.Methods:Carotid plaque cases were independently collected by two sonographers as a training set and a validation set.In the training set and validation set,patients were divided into symptomatic group and asymptomatic group according to whether they had corresponding cerebral ischemia symptoms recently.The general clinical data of the two groups were compared.The radiomics features of carotid plaque based on gray scale ultrasound images were extracted.In the training set,with the presence or absence of cerebral ischemia as the dependent variable and the plaque imaging features as the independent variable,LASSO regression was used to screen the independent variables to construct the radiomics model.The ROC curve,calibration curve and decision curve were used to verify the model internally and externally.Results:There was no significant difference in the general clinical data between the cerebral infarction group and the non-cerebral infarction group in the training set and the verification set(P＞0.05).In the training set,six radiomics features were selected to build the model.In the training set,the AUC and C-index of the model were 0.913[95％CI(0.862-0.964)]and 0.887.In the validation set,the AUC was 0.837[95％CI(0.758-0.917)]and the C-index was 0.798.In the training set and verification set,the calibration curve fits the ideal curve well,and the DCA curve has high net benefit value.Conclusion:The radiomics model based on gray-scale ultrasound is feasible and valuable to identify symptomatic carotid plaque.