Clinical Scorecard: A Systematic Review and Evaluation of Computer Vision Technologies in Vascular Surgery
At a Glance
Category
Detail
Condition
Vascular diseases including aortic pathologies, carotid stenosis, foot ulcers, and peripheral artery disease
Key Mechanisms
Application of computer vision techniques for image analysis, segmentation, and classification in vascular surgery
Target Population
Patients with vascular diseases such as aortic aneurysms, carotid artery stenosis, diabetic foot ulcers, and peripheral artery disease
Care Setting
Vascular surgery clinical and research settings utilizing imaging data
Key Highlights
288 studies reviewed with rapid growth in computer vision applications since 2017
Major focus on aortic pathologies (33%), carotid stenosis (30%), and foot ulcers (25%), with limited research on peripheral artery disease (6%)
Commonly reported performance metrics were Dice coefficient (51%) and accuracy (36%), with infrequent use of AUROC (17%)
Guideline-Based Recommendations
Diagnosis
Use Dice coefficient for evaluating segmentation tasks
Use AUROC for discrimination tasks in diagnostic models
Management
Incorporate computer vision tools to support clinical decision-making in vascular surgery
Focus development on underrepresented conditions such as peripheral artery disease
Monitoring & Follow-up
Adhere to TRIPOD+AI guidelines early during model development to improve transparency and reduce bias
Risks
High risk of bias in majority of studies (85%) necessitates cautious interpretation
Poor adherence (57%) to reporting standards may limit reproducibility and clinical translation
Patient & Prescribing Data
Patients with vascular diseases undergoing imaging evaluation
Computer vision models can assist in accurate image segmentation and classification to inform treatment planning, but require rigorous validation and standardized reporting
Clinical Best Practices
Consult TRIPOD+AI statement during early stages of model development
Prioritize use of standardized performance metrics such as Dice coefficient and AUROC depending on task
Expand research focus to include peripheral artery disease to address current gaps
Ensure prospective data collection and clinical trials to validate computer vision applications
Maintain critical appraisal of bias risk and methodological quality in studies
The Vascular Surgery Fellowship, which is accredited by the Accreditation Council for Graduate Medical Education (ACGME), is a program of Cedars-Sinai, an academic medical center renowned for vascular care, research and education. Cedars-Sinai is consistently ranked one of the nation's best hospitals by U.S. News & World Report.
Researchers compare personalized versus standard prehabilitation and examine functional, immune, and postoperative outcomes before major elective surgery.
