Related Experiment Video
Updated: Apr 18, 2026
04:47
Author Spotlight: Learning Systematic Bronchoscopy in a Simulation-Base Setting
Published on: June 23, 2023
4.0K
An autonomous bronchoscopy robot controlled by artificial intelligence (BronchoBot) outperforms experienced
Kristoffer M Cold1,2, Zhuoqi Cheng3, Lars Konge1,4
1Copenhagen Academy for Medical Education and Simulation (CAMES), Rigshospitalet, University of Copenhagen and the Capital Region of Denmark, Copenhagen, Denmark.
ERJ Open Research
|April 17, 2026
Summary
An autonomous bronchoscopy robot, BronchoBot, demonstrated superior performance compared to human bronchoscopists in a simulation. BronchoBot achieved higher ratings in anatomy and dexterity, and improved diagnostic completeness and procedure efficiency.
Area of Science:
- Medical Robotics
- Endoscopy Technology
- Surgical Simulation
Background:
- Flexible bronchoscopy performance varies significantly among practitioners.
- There is a need for standardized and objective assessment of bronchoscopy skills.
Purpose of the Study:
- To compare the performance of an autonomous bronchoscopy robot (BronchoBot) against human bronchoscopists.
- To evaluate BronchoBot's efficacy in a standardized simulation-based setting.
Main Methods:
- A realistic phantom was used for 10 BronchoBot procedures and 30 human procedures (novice, intermediate, experienced).
- Primary outcomes included Anatomy and Dexterity Ratings by blinded experts.
- Secondary outcomes (Diagnostic Completeness, Structured Progress, Procedure Time, Mean Intersegmental Time) were assessed by artificial intelligence (AI).
Main Results:
- BronchoBot significantly outperformed all human groups across all measured outcomes.
- BronchoBot achieved higher Anatomy (11.8 vs. 10.3) and Dexterity (5.3 vs. 4.5) ratings than experienced humans.
- BronchoBot demonstrated superior Diagnostic Completeness (18 vs. 16 segments) and efficiency (124s vs. 190s Procedure Time).
Conclusions:
- BronchoBot surpassed experienced bronchoscopists in a simulated environment based on expert and AI evaluations.
- The autonomous system shows potential for benchmarking expert performance in bronchoscopy.
- Further research and clinical evaluation are necessary before patient application.
