Impact of Artificial Intelligence on the Timing of Recurrent Laryngeal Nerve Recognition during Robot-Assisted Minimally Invasive Esophagectomy

Tasuku Furube ¹, Masashi Takeuchi ², Hirofumi Kawakubo ³

⁰Department of Surgery, Keio University School of Medicine, Tokyo, Japan.

Annals of Surgical Oncology +

|

June 26, 2025

View abstract on PubMed

Summary

Artificial intelligence (AI) significantly speeds up recurrent laryngeal nerve (RLN) identification during robot-assisted minimally invasive esophagectomy (RAMIE). This AI tool enhances surgeon confidence in recognizing the RLN, crucial for preventing nerve palsy.

Area Of Science

Robotic Surgery
Surgical Navigation
Artificial Intelligence in Medicine

Background

Recurrent laryngeal nerve (RLN) palsy is a risk in esophagectomy.
An AI-based anatomical recognition system was developed for robot-assisted minimally invasive esophagectomy (RAMIE).
The study aimed to assess AI's ability to expedite RLN recognition by surgeons.

Purpose Of The Study

To evaluate the impact of an AI system on the speed and confidence of recurrent laryngeal nerve (RLN) identification during RAMIE.
To establish a confidence level (CL) system for RLN recognition and measure its change with AI assistance.

Main Methods

Five RAMIE surgical videos were analyzed using an AI system.
A confidence level (CL) scale (CL0, CL1, CL2) was used to quantify RLN recognition stages.
Eight trainee surgeons compared RLN identification times with and without AI assistance over >4 weeks.

Main Results

AI significantly reduced the time to identify both right and left recurrent laryngeal nerves (RLNs).
For right RLN recognition, AI decreased time to CL1 (134s vs 178s) and CL2 (233s vs 325s) (p < 0.001).
AI usage led to faster RLN identification across all tested scenarios.

Conclusions

AI enables surgeons to rapidly identify the recurrent laryngeal nerve (RLN) during RAMIE.
The AI system enhances surgeon confidence in RLN identification.
AI is a valuable tool for preventing RLN palsy in complex surgeries.