Related Experiment Video
Updated: Jun 10, 2025
12:33
Corticospinal Excitability Modulation During Action Observation
Published on: December 31, 2013
8.8K
Detection of hand motion during cadaveric mastoidectomy dissections: a technical note
Thomas J On1, Yuan Xu1, Nicolas I Gonzalez-Romo1
1The Loyal and Edith Davis Neurosurgical Research Laboratory, Department of Neurosurgery, Barrow Neurological Institute, St. Joseph's Hospital and Medical Center, Phoenix, AZ, United States.
Frontiers in Surgery
|October 18, 2024
Summary
A deep learning model tracked surgeon hand motions during cadaveric mastoidectomies, showing potential for refining surgical precision. Further validation is needed to use these metrics for surgical training and evaluation.
Area of Science:
- Neurosurgery
- Surgical Technology
- Artificial Intelligence in Medicine
Background:
- Posterior temporal bone access demands precise drilling to avoid critical structure injury.
- Surgical skill and precision are paramount in complex procedures like mastoidectomy.
Purpose of the Study:
- To evaluate a deep learning hand motion detector for precision during power drill use in cadaveric mastoidectomy.
- To assess the feasibility of using AI to refine surgical hand movements and provide performance metrics.
Main Methods:
- A deep learning model tracked 21 hand landmarks during three cadaveric mastoidectomies.
- Horizontal and vertical motion tracking plots were generated from landmark data.
- Preliminary surgical performance metrics were derived from motion detection data.
Main Results:
- Over 1.9 million landmark detections were analyzed with 85.9% overall performance.
- Hand motion detection showed balanced performance between dominant (48.2%) and non-dominant (51.7%) hands.
- Tracking challenges included microscope light glare and hands moving out of the camera's field of view.
Conclusions:
- A sensor-free deep learning system can measure surgical hand motion during mastoidectomy simulations.
- Preliminary metrics for assessing hand motion were developed, requiring further validation for surgical training applications.
