Utilizing artificial intelligence (AI) technology for the segmentation of plaques on ultrasound images to evaluate the stability of carotid artery plaques and analyze its diagnostic accuracy in differentiating vulnerable plaques from stable ones. A retrospective study was conducted on 202 patients with ischemic stroke, who were divided into vulnerable plaque group (85 cases) and stable plaque group (117 cases) based on the results of carotid color Doppler ultrasound examination. From the vulnerable plaque group, 63 cases were randomly selected as the modeling group and 22 cases as the validation group; similarly, from the stable plaque group, 87 cases were randomly selected as the modeling group and 30 cases as the validation group. Based on the ultrasound images of the modeling group, plaques were segmented using artificial intelligence technology, and 1414 radiomics features were extracted. These features were then subjected to dimensionality reduction and feature selection using the least absolute shrinkage and selection operator (LASSO) method. Subsequently, a Support Vector Machine (SVM) model was constructed and validated using the selected features. The sensitivity, specificity, and Area Under the Curve (AUC) of the model were evaluated through the analysis of the receiver operating characteristic (ROC) curve. A total of 43 radiomics feature parameters were selected by the LASSO method. The training group for the SVM model had an AUC of 89.42% (95% CI: 80.74-98.10%), sensitivity of 79.84%, and specificity of 93.10%, while the validation group had an AUC of 82.73% (95% CI: 71.64-93.81%), sensitivity of 81.82%, and specificity of 80.00%. The use of artificial intelligence technology for the segmentation of plaques in ultrasound images, coupled with the analysis of radiomics models, can efficiently distinguish the stability of carotid artery plaques, providing a diagnostic basis for the clinical prediction of ischemic stroke. Not applicable.

Background: While conventional ultrasound (US)-guided core needle biopsy (CNB) is routinely used for preoperative diagnosis of breast lesions, it is susceptible to false negatives and underestimation, especially in B1–B3 categories, making accurate diagnosis a significant clinical challenge. This study aimed to explore the conventional US and contrast-enhanced ultrasound (CEUS) features that predict postoperative malignancy upgrading of category B1–B3 breast lesions diagnosed by CEUS-guided CNB. Methods: A total of 84 breast lesions that underwent US, CEUS, CEUS-guided CNB and surgery were collected from February 2023 to February 2025, retrospectively. The malignancy upgrading was defined as CNB showing a B1–B3 classification, but surgical pathology showing a B5 classification. Uni-variable analysis compared US and CEUS features between non-upgrading and upgrading groups, and significant features were entered into multi-variable logistic regression. The performances of each independent predictor and their combinations were evaluated using receiver operating characteristic curve analysis. Results: Thirteen B1–B3 lesions were upgraded to B5 lesions. The multivariate analysis showed that size >20 mm on conventional US [odds ratio (OR) =5.85; 95% confidence interval (CI): 1.49–22.92; P=0.011], and enlarged enhancement range on CEUS (OR =17.44; 95% CI: 2.03–150.02; P=0.009) were independent predictors. The multi-variable model incorporating these two features yielded an area under the curve (AUC) of 0.83 and showed significantly higher predictive efficiency than size >20 mm (AUC =0.68) and enlarged enhancement range (AUC =0.74) (P=0.027 and P=0.013, respectively). Conclusions: Both conventional US and CEUS can provide useful information for postoperative malignancy upgrading in B1–B3 lesions. The integration of them improved the predictive performance.