Operational effectiveness of three-dimensional flexible endoscopy: an ex vivo study using a new model

Two-dimensional (2D) images lack depth information and thus provide probabilistic recognition that do not completely match the actual three-dimensional (3D) information. Here, we investigated the operability of 3D endoscopes.

A 3D operation model was developed by passing 20 silk threads through upper and lower plates at 2-mm intervals in front and back rows separated by 1 mm. We evaluated accuracy and time of operating an electrosurgical knife. A successful operation was defined as pulling only a front-row thread; an unsuccessful operation was defined as pulling no thread (miss) or simultaneously pulling front- and back-row threads. Endoscopists (four experts, six trainees) repeated the operation under 2D and 3D conditions until individually accumulating 10 successful attempts under each condition.

Operation accuracy was significantly higher for 3D compared with 2D in all endoscopists (88.5% vs. 61.3%; p < 0.01) and in both experience groups (trainees: 84.5% vs. 61.2%; experts: 95.2% vs. 61.5%; both p < 0.01). Operation time was significantly shorter for 3D compared with 2D in all endoscopists (12.5 ± 4.1 s vs. 14.8 ± 4.7 s; p < 0.01) and in both experience groups (trainees: 12.8 ± 4.2 s vs. 15.2 ± 4.9 s; experts: 12.1 ± 4.0 s vs. 14.3 ± 4.3 s; both p < 0.01).

Compared with 2D endoscopy, 3D endoscopy significantly improved operation accuracy and shortened operation time, suggesting that 3D endoscopy enables accurate operation by depth information, aiding spatial recognition.