Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: May 28, 2026

A Training and Testing System for Performing Vascular Reconstruction In Vitro

A Training and Testing System for Performing Vascular Reconstruction In Vitro

Published on: October 26, 2019

Machine Learning Models for Objective Assessment of Vascular Anastomoses Using Computational Fluid Dynamics for

Levente Kiss-Pápai¹, Stefánia Reich^1,2, Júlia Varga^1,2

¹Institute of Transdisciplinary Discoveries, University of Pécs, Szigeti út 12., 7624 Pécs, Hungary.

Journal of Clinical Medicine

|May 27, 2026

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Interplay of the ENS and Microbiota With Murine Gut Epithelium-Derived Organoids in Aging.

Aging cell·2026

Same author

Microsurgical treatment of lower extremity lymphedema: An umbrella systematic review supporting the American Venous Forum and the American Vein and Lymphatic Society clinical practice guidelines on management of lymphedema.

Journal of vascular surgery. Venous and lymphatic disorders·2026

Same author

The continuing challenge of Nutcracker syndrome.

Journal of vascular surgery. Venous and lymphatic disorders·2026

Same author

Nonmyeloablative pentostatin-cyclophosphamide preconditioning improves rates of engraftment in adults undergoing haploidentical HCT for sickle cell disease.

PloS one·2026

Same author

Contemporary management of superior vena cava syndrome.

Journal of vascular surgery. Venous and lymphatic disorders·2026

Same author

Risk-Adjusted Outcomes After Minimally Invasive Direct Coronary Artery Bypass Grafting: A Multicentre Experience.

European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery·2026

Machine learning regression models accurately predict surgical skill in vascular anastomoses using computational fluid dynamics features. This approach offers reproducible, mechanistically relevant feedback for surgical training.

Area of Science:

Biomedical Engineering
Computational Fluid Dynamics
Machine Learning

Background:

Objective surgical skill assessment is crucial but current methods are labor-intensive and focus on the trainee, not the outcome.
Machine learning (ML) offers reproducible feedback but often uses limited data types, reducing assessment to simple classifications.
Assessing the final product of vascular anastomoses using ML and computational fluid dynamics (CFD) features is an underexplored area.

Purpose of the Study:

Compare ML regression models for predicting expert scores of vascular anastomoses based on CFD features.
Evaluate the biomechanical plausibility of ML predictions in surgical skill assessment.
Introduce a novel method for objective and reproducible feedback in vascular surgical training.

Main Methods:

Keywords:

machine learning performance assessment skill training surgical education vascular surgery

More Related Videos

Development of a Murine Model for Femoral Artery Anastomotic Stenosis

Development of a Murine Model for Femoral Artery Anastomotic Stenosis

Published on: April 18, 2025

Related Experiment Videos

Last Updated: May 28, 2026

A Training and Testing System for Performing Vascular Reconstruction In Vitro

A Training and Testing System for Performing Vascular Reconstruction In Vitro

Published on: October 26, 2019

Development of a Murine Model for Femoral Artery Anastomotic Stenosis

Development of a Murine Model for Femoral Artery Anastomotic Stenosis

Published on: April 18, 2025

146 participants created 419 end-to-side anastomoses on 3D printed simulators.
Anastomoses were 3D scanned, expert-ranked, and analyzed using CFD-derived hemodynamic features.
Various ML models, including XGBoost, were trained and validated using nested cross-validation with hyperparameter optimization.

Main Results:

Expert ranking showed strong inter-rater reliability (ICC3k = 0.846).
XGBoost demonstrated superior performance with the lowest RMSE (0.758) and highest R² (0.673).
Key CFD features like wall shear stress gradient and maximum pressure were identified as influential predictors.

Conclusions:

Tree-based ensemble methods, especially XGBoost, effectively model biomechanical properties against expert scores.
Combining CFD and ML provides reproducible and mechanistically relevant feedback for vascular surgical skill development.
This approach enhances objective performance assessment in surgical training by focusing on the procedure's end-product.