Related Experiment Video
Updated: May 31, 2026
Robotized Testing of Camera Positions to Determine Ideal Configuration for Stereo 3D Visualization of Open-Heart Surgery
Published on: August 12, 2021
Quantitative assessment of laparoscopic camera navigation skill: a retrospective cross-sectional study
Annika Lindstrom Haughey1, Shannon Barter2, Cameron Reid3
1Department of Mechanical Engineering, Duke University, MEMS Dept. Box 90300, Durham, NC, 27708, USA. annika.haughey@duke.edu.
Objective:
To evaluate whether a low-cost, automated tool using trajectory-based metrics and real-time computer vision can objectively assess laparoscopic camera navigation (LCN) skill and distinguish between levels of surgical experience. Laparoscopic camera navigation (LCN) is essential to the safety and success of minimally invasive surgery. Despite its importance, LCN is often overlooked in formal training and lacks standardized assessment methods. Existing tools are labor-intensive, subjective, and not scalable, creating a need for efficient, objective evaluation strategies.
Methods:
This cross-sectional observational validation study was conducted from January to April 2025 in a single-site academic surgical training laboratory. Fifty-nine participants (6 expert surgeons, 12 senior residents, 26 junior residents, and 15 novices) completed a standardized LCN task using a custom-designed maze and a 30° laparoscope in a bench-top trainer. Real-time computer vision detected targets, while an electromagnetic tracking system recorded the laparoscope's trajectory. Five automated performance metrics were collected: space coverage, dimensionless jerk, rotational smoothness, idle time, and total task time. A principal component analysis (PCA) was used to generate a composite performance score (0-100).
Results:
All participants completed the task. Experts demonstrated lower space coverage (121 ± 31 cm2) and shorter completion times (163 ± 59 s) compared with novices (205 ± 67 cm2; 497 ± 133 s). Experts also outperformed novices in jerk, rotational smoothness, and idle time metrics. PCA-derived scores were significantly higher for experts (92 ± 3) than novices (34 ± 24), with all metrics showing statistically significant differences (p < .005).
Conclusions:
This automated, low-cost tool reliably quantified LCN skill across training levels using interpretable metrics. Its objectivity and scalability support integration into surgical education for formative feedback and standardized assessment.
