Evaluation of the Effectiveness of Neural Network Models for Automated Classification of Surgical Suture Quality on Limited Datasets
https://doi.org/10.46594/2687-0037_2025_3_2127
Abstract
The results of the analysis of the effectiveness of neural networks for automatic classification of the quality of surgical sutures based on photographs are presented. High accuracy (F1-measure > 0.90) was achieved for nodular, vascular, and laparoscopic sutures on small datasets (100-190 images). The technology is promising for an objective assessment of surgical skills.
About the Authors
R. V. Ishchenko
Институт неотложной и восстановительной хирургии им. В. К. Гусака
Russian Federation
Russian Federation
M. V. Solopov
V. V. Turchin
A. G. Popandopulo
O. S. Antonyuk
A. A. Ermak
K. K. Ladyk
F. S. Popivnenko
K. O. Golubitsky
A. E. Glebova
D. A. Filimonov
Review
For citations:
Ishchenko R.V., Solopov M.V., Turchin V.V., Popandopulo A.G., Antonyuk O.S., Ermak A.A., Ladyk K.K., Popivnenko F.S., Golubitsky K.O., Glebova A.E., Filimonov D.A. Evaluation of the Effectiveness of Neural Network Models for Automated Classification of Surgical Suture Quality on Limited Datasets. Virtual Technologies in Medicine. 2025;(3):328-329. (In Russ.) https://doi.org/10.46594/2687-0037_2025_3_2127
Views:
4
Email this article 