Video-Based Detection and Analysis of Errors in Robotic Surgical Training

Robot-assisted minimally invasive surgeries offer many advantages but require complex motor tasks that take surgeons years to master. There is currently a lack of knowledge on how surgeons acquire these robotic surgical skills. To help bridge this gap, we previously followed surgical residents learning complex surgical training dry-lab tasks on a surgical robot over six months. Errors are an important measure for self-training and for skill evaluation, but unlike in virtual simulations, in dry-lab training, errors are difficult to monitor automatically. Here, we analyzed the errors in the ring tower transfer task, in which surgical residents moved a ring along a curved wire as quickly and accurately as possible. We developed an image-processing algorithm to detect collision errors and achieved detection accuracy of ~95%. Using the detected errors and task completion time, we found that the surgical residents decreased their completion time and number of errors over the six months. This analysis provides a framework for detecting collision errors in similar surgical training tasks and sheds light on the learning process of the surgical residents.

@article{lev2025_2504.19571,
  title={ Video-Based Detection and Analysis of Errors in Robotic Surgical Training },
  author={ Hanna Kossowsky Lev and Yarden Sharon and Alex Geftler and Ilana Nisky },
  journal={arXiv preprint arXiv:2504.19571},
  year={ 2025 }
}