BPG is committed to discovery and dissemination of knowledge
Minireviews
Copyright ©The Author(s) 2025.
World J Radiol. Sep 28, 2025; 17(9): 110447
Published online Sep 28, 2025. doi: 10.4329/wjr.v17.i9.110447
Table 1 Summary of artificial intelligence-based carotid computed tomography angiography plaque detection studies
Ref.
Sample size
Model type
Dataset source
Results
Clinical impact
Jie et al[6]3245 (meta-analysis, 17 studies)Mixed AI modelsMulticenter/multi-countrySensitivity of 0.91, specificity of 0.88, AUC of 0.94Improved CTA plaque detection accuracy
Pisu et al[7]156MLSingle-center CTAAUC of 0.91, sensitivity of 87%, specificity of 85%Early identification of high-risk symptomatic plaques
Shi et al[8]112Radiomics + logistic regressionSingle-center CTAAUC of 0.89Differentiates plaque stability, optimizes treatment planning
Zhai et al[9]1234Convolutional neural network (fully automatic detection)Multicenter CTASensitivity of 0.93, specificity of 0.92Rapid automated plaque screening
Hu et al[10]205Radiomics + dual-energy CTAMulticenter CTAAUC of 0.90, accuracy of 88%Improved symptomatic plaque recognition
Guo et al[11]120Two-stage deep learningSingle-center CTASensitivity of 0.92, specificity of 0.89Good performance in early-stage validation
Xie et al[12]560Swin-UNet + multi-scale supervisionMulticenter CTADice coefficients of 0.93High-precision segmentation for quantitative analysis
Song et al[13]98AI segmentation + ultrasound radiomicsSingle-center ultrasoundAUC of 0.88Ultrasound-based method applicable to CTA risk assessment
Wei et al[14]4562ML (random forest, etc.)Single-center health check-upAUC of 0.85Early population risk prediction tool