Does Deep Learning Vascular Segmentation on CTA Improve Vertebral Artery Dissection Detection?
- 1Duke University
Abstract
PurposeThe aims of this study were to develop an externally validated deep learning vascular segmentation model for digital enhancement of vertebral artery dissection detection on CTA and to assess clinical utility of the model through a paired crossover reader study.
Materials and MethodsThis retrospective, IRB-approved study conducted from September 2024 to July 2026 included an internal training cohort of 84 manually segmented CTAs plus 17 CTAs from the RSNA Intracranial Aneurysm Challenge (101 CTAs total) and an external testing cohort of 40 CTAs (22 positive, 18 negative for vertebral artery dissection). A nnU-Net (version 2) model was trained using five-fold cross-validation with the entire training cohort. Technical performance was evaluated using the Dice similarity coefficient (DSC). A two-part crossover reader study included 10 readers who assessed diagnostic accuracy, confidence, and interpretation time with and without segmentation-based augmentation. Statistical analyses included McNemars exact test (accuracy) and Wilcoxon signed-rank test (confidence, time), with P < .05 considered significant.
ResultsThe model achieved a DSC of 0.96 {+/-} 0.01. In the reader study, diagnostic accuracy was lower with augmentation than without (0.72 vs 0.84, P < 0.001), while confidence (3.98 vs 3.96, P = .77) and interpretation time (137.3 vs 166.7 seconds, P = .54) did not differ significantly. Subjectively, 7 of 10 readers reported they would use the tool routinely in acute or trauma settings.
ConclusionAlthough the deep learning model demonstrated accurate segmentation of vascular structures, further validation is needed to establish its utility for enabling accurate diagnosis of VAD in fast-paced clinical settings.
Key Points- A deep learning vascular segmentation model trained for Vertebral Artery Dissection achieved a Dice similarity coefficient of 0.96 {+/-} 0.01 on external data. - Diagnostic accuracy on a follow-up reader study was significantly lower on CTA from the segmentation model compared to conventional CTA (0.72 vs 0.84, p < 0.001). - Seven out of ten radiology trainees and attendings reported they would routinely use digitally enhanced CTA in acute or trauma settings.
Summary StatementOn direct comparison between conventional CTA and CTA enhanced with deep learning vascular segmentation, radiology trainees and attendings were significantly more accurate at diagnosing vertebral artery dissection on conventional CTA, with no significant difference in interpretation time or confidence.
Related
- Osseous and Vascular Course Anomalies in Nontraumatic Vertebral Artery Dissection: A Per-Vessel Case-Control Study
- Deep learning model for simultaneous vertebral level localization and fracture detection on DXA vertebral fracture assessment images
- Reliability of the L3 Transverse Process as a Landmark for Lumbar Spine Enumeration Across Different Vertebral Segmentation Patterns: A Retrospective CT Study
- Ischemic Stroke Associated With Vertebral Artery Occlusion Following Cervical Spine Manipulation in a Young Woman With Underlying Vascular Risk Factors: A Case Report.
- Deep learning based edge segmentation and classification for plasterboard defect detection
BibTeX
@misc{zaveri2026does,
title = {{Does Deep Learning Vascular Segmentation on CTA Improve Vertebral Artery Dissection Detection?}},
author = {S. Zaveri and D. Zhang and E. Castellino and E. Calabrese},
howpublished = {medRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.20.26363486},
url = {https://www.medrxiv.org/content/10.64898/2026.09.20.26363486v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Does Deep Learning Vascular Segmentation on CTA Improve Vertebral Artery Dissection Detection?
AU - S. Zaveri
AU - D. Zhang
AU - E. Castellino
AU - E. Calabrese
PY - 2026
DA - 2026/09/20
PB - medRxiv
DO - 10.64898/2026.09.20.26363486
UR - https://www.medrxiv.org/content/10.64898/2026.09.20.26363486v1
AB - PurposeThe aims of this study were to develop an externally validated deep learning vascular segmentation model for digital enhancement of vertebral artery dissection detection on CTA and to assess clinical utility of the model through a paired crossover reader study. Materials and MethodsThis retrospective, IRB-approved study conducted from September 2024 to July 2026 included an internal training cohort of 84 manually segmented CTAs plus 17 CTAs from the RSNA Intracranial Aneurysm Challenge (101 CTAs total) and an external testing cohort of 40 CTAs (22 positive, 18 negative for vertebral artery dissection). A nnU-Net (version 2) model was trained using five-fold cross-validation with the entire training cohort. Technical performance was evaluated using the Dice similarity coefficient (DSC). A two-part crossover reader study included 10 readers who assessed diagnostic accuracy, confidence, and interpretation time with and without segmentation-based augmentation. Statistical analyses included McNemars exact test (accuracy) and Wilcoxon signed-rank test (confidence, time), with P < .05 considered significant. ResultsThe model achieved a DSC of 0.96 {+/-} 0.01. In the reader study, diagnostic accuracy was lower with augmentation than without (0.72 vs 0.84, P < 0.001), while confidence (3.98 vs 3.96, P = .77) and interpretation time (137.3 vs 166.7 seconds, P = .54) did not differ significantly. Subjectively, 7 of 10 readers reported they would use the tool routinely in acute or trauma settings. ConclusionAlthough the deep learning model demonstrated accurate segmentation of vascular structures, further validation is needed to establish its utility for enabling accurate diagnosis of VAD in fast-paced clinical settings. Key Points- A deep learning vascular segmentation model trained for Vertebral Artery Dissection achieved a Dice similarity coefficient of 0.96 {+/-} 0.01 on external data. - Diagnostic accuracy on a follow-up reader study was significantly lower on CTA from the segmentation model compared to conventional CTA (0.72 vs 0.84, p < 0.001). - Seven out of ten radiology trainees and attendings reported they would routinely use digitally enhanced CTA in acute or trauma settings. Summary StatementOn direct comparison between conventional CTA and CTA enhanced with deep learning vascular segmentation, radiology trainees and attendings were significantly more accurate at diagnosing vertebral artery dissection on conventional CTA, with no significant difference in interpretation time or confidence.
ER -