EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep learning algorithm enables lower limb venous thrombosis detection with CT venography

Shanshan Shen, Haixiao Yang, Pengchao Wang, Jing Zhang, Jiahao Zhen, Tao Liu

Frontiers in Cardiovascular Medicine · 12.08.2026 · 13

DOI: 10.3389/fcvm.2026.1890645 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Cardiovascular Medicine, online erschienen am 12.08.2026. Verfasst von Shanshan Shen und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Hebei General Hospital, Shijiazhuang, Hebei, China. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Lower limb computed tomography venography (CTV) has low success rates for deep vein thrombosis (DVT) diagnosis. This study applied deep learning to improve DVT identification.

Methods

Our study enrolled 119 positive DVT and 40 negative DVT, 111 positive pulmonary embolism (PE) patients and 20 negative PE. Two algorithms were evaluated: Faster R-CNN trained directly on CTV images, and YOLO11-nano pre-trained on computed tomographic pulmonary angiography (CTPA) then optimized on CTV via transfer learning. Three radiologists (3-5 years' experience) independently interpreted CTV images. Diagnostic performance was compared.

Results

The YOLO11-nano model and Faster R-CNN achieved over 97% accuracy and sensitivity in diagnosing PE. Compared to the three radiologists in diagnosing DVT, YOLO also had an excellent accuracy (92.1%) and sensitivity (93.9%), superior to Faster R-CNN (66% accuracy, 68% sensitivity). The YOLO model exceeded 77% accuracy in pelvic and femoral-popliteal segments and 66.7% in detecting calf segment thrombi.

Discussion

The CTPA-based transfer learning model significantly improved the feasibility of this method in routine CTV diagnostic performance, offering a promising approach for simultaneous PE and DVT detection.

Schlagwörter

Stichwörter der Autor:innen

CT venographyDVTdeep learningpulmonary embolismtransfer learning

Autor:innen und Institutionen

  • Department of Radiology, Hebei General Hospital, Shijiazhuang, Hebei, China.
  • Department of Radiology, The First Hospital of Hebei Medical University, Shijiazhuang, Hebei, China.
  • School of Economics and Management, Hebei University of Science and Technology, Shijiazhuang, Hebei, China.

Zitieren

APA

Shen, S., Yang, H., Wang, P., Zhang, J., Zhen, J., & Liu, T. (2026). Deep learning algorithm enables lower limb venous thrombosis detection with CT venography. Frontiers in Cardiovascular Medicine, 13. https://doi.org/10.3389/fcvm.2026.1890645

Vancouver

Shen S, Yang H, Wang P, Zhang J, Zhen J, Liu T. Deep learning algorithm enables lower limb venous thrombosis detection with CT venography. Frontiers in Cardiovascular Medicine. 2026;13. doi:10.3389/fcvm.2026.1890645

BibTeX
@article{Shen2026,
  author = {Shen, Shanshan and Yang, Haixiao and Wang, Pengchao and Zhang, Jing and Zhen, Jiahao and Liu, Tao},
  title = {Deep learning algorithm enables lower limb venous thrombosis detection with CT venography},
  journal = {Frontiers in Cardiovascular Medicine},
  year = {2026},
  volume = {13},
  doi = {10.3389/fcvm.2026.1890645}
}
Kostenlos starten →