Key result
Multitask deep learning improves carotid plaque detection to ~0.86 AUC over baseline models.
Why the study?
Do multitask deep learning models with anatomical auxiliary segmentation tasks (vessel lumen, IMT, or plaque) improve carotid plaque detection performance compared to a classification-only baseline on carotid ultrasound in an external validation cohort?
Cohort (n=525)
Yes
Do multitask deep learning models with anatomical auxiliary segmentation tasks (vessel lumen, IMT, or plaque) improve carotid plaque detection performance compared to a classification-only baseline on carotid ultrasound in an external validation cohort?
Incorporating anatomical auxiliary segmentation tasks into deep learning models for carotid ultrasound significantly improves plaque detection over classification-only models and allows tailoring of sensitivity versus specificity trade-offs for screening or confirmatory clinical scenarios.
Answers are grounded in this paper and cite their sources.
Should not yet change carotid plaque detection practices; leaves open auxiliary task optimization for future deep learning models.
Purpose Carotid plaque screening using ultrasound is challenged by class imbalance and ambiguous lesion boundaries. However, how different anatomical auxiliary tasks influence carotid plaque detection performance and clinical operating characteristics remains insufficiently investigated. We aimed to externally evaluate multitask deep learning models with different anatomical auxiliary tasks and investigate their effects on clinical performance profiles. Methods We developed a multitask network incorporating three anatomical auxiliary segmentation targets (vessel lumen, intima-media thickness (IMT), and plaque), together with a classification-only baseline. The model was developed on 1297 images from 262 patients and evaluated on an external cohort of 1091 images from 263 patients. To ensure fair comparison, we froze the encoder of each variant and retrained an identical classification head on the internal training set, and then applied a fixed threshold determined via Youden index maximization for each model on the internal validation set to the external cohort. Results In external validation (263 patients, 94 plaque-positive), all multitask models outperformed the classification-only baseline (AUC 0.615, p < 0.0001), while showing comparable AUCs among themselves (lumen-guided 0.852, IMT-guided 0.853, plaque-guided 0.858; all p > 0.76). Despite comparable discrimination, they exhibited distinct clinical profiles: the IMT-guided model achieved the highest sensitivity (SEN 0.830), the lumen-guided model achieved the highest specificity (SPE 0.864), and the plaque-guided model offered more balanced performance (SEN 0.723, SPE 0.822). Radiofrequency balloon ablation significantly reduced external AUC for the lumen-guided (0.852 vs 0.790; p=0.0145) and plaque-guided (0.858 vs 0.796; p=0.0091) models, but not for the IMT-guided model (p=0.2171), indicating task-dependent effects. Conclusion The choice of auxiliary task does not confer universal superiority in discrimination but instead determines the model's clinical orientation, allowing task-specific adaptation to different screening or confirmatory scenarios.
No takes yet. Share an insight, caveat, or question.
施 et al. (2026) conducted a cohort in Carotid plaque (n=525). Multitask deep learning models with anatomical auxiliary segmentation tasks (vessel lumen, intima-media thickness, and plaque) vs. Classification-only baseline deep learning model was evaluated on Carotid plaque detection discrimination (AUC) in the external validation cohort (p=<0.0001). Multitask deep learning models with anatomical auxiliary tasks outperformed the classification-only baseline for carotid plaque detection (AUC 0.852-0.858 vs 0.615, p<0.0001).
Synapse has enriched 10 related papers. Consider them for comparative context: