Contrast-enhanced ultrasound based radiomics model enhances identification of symptomatic carotid plaque

Background: We aimed to identify the most valuable variables in multi-modal ultrasound technologies and compare the performance of radiomics models in detecting symptomatic carotid plaques. Methods: From August 2013 to April 2021, we retrospectively enrolled 112 carotid plaque patients and randomly divided them into training and validation cohorts at a 3:1 ratio. A traditional model (T model) was built using independently significant variables from multivariate analysis of clinical and multi-modal ultrasound characteristics. Radiomics models based on conventional ultrasound, contrast-enhanced ultrasound (CEUS), and a combined approach (USR, CEUSR, and US-CEUSR) were then developed. Combined with T model, three incorporated models (TUSR, TCEUSR, and TUS-CEUSR) were established. Model performance was assessed using area under the receiver operating characteristic curve (AUC) and decision curve analysis. Results: Only intraplaque contrast enhancement from CEUS modality was independently associated with clinical symptoms [odds ratio = 3.750; 95% confidence interval (CI), 1.403–10.881; p = 0.010] and was used to construct the T model, which achieved an AUC of 0.583 (95% CI, 0.390–0.776) in the validation cohort. The CEUSR model had an AUC of 0.813 (95% CI, 0.626–0.957), surpassing the T model ( p < 0.001). The USR model's AUC was 0.663 (95% CI, 0.427–0.866). Notably, the CEUSR model outperformed the USR model ( p = 0.043). Other integrated models didn't significantly surpass the CEUSR model. Conclusion: Intraplaque contrast enhancement, a qualitative variable by CEUS, is an independent risk factor for symptomatic carotid plaques, and radiomics features from CEUS can effectively predict symptomatic carotid plaques.