Gasless transaxillary endoscopic thyroidectomy (GTET), as a minimally invasive surgical approach, achieves scarless neck outcomes through axillary incisions. While demonstrating superior cosmetic advantages, this technique presents considerable technical challenges due to its anatomical complexity and elevated risk of iatrogenic injury to critical structures such as the recurrent laryngeal nerve (RLN). Notably, a comparative analysis of minimally invasive thyroidectomy approaches revealed the highest incidence of RLN injury associated with GTET procedures.1 Recent advancements in artificial intelligence (AI)-assisted intraoperative navigation systems have shown significant potential in addressing these challenges. Emerging evidence indicates that AI-powered neural recognition algorithms can enhance RLN identification accuracy, effectively reducing intraoperative nerve injury rates and improving overall procedural safety parameters.2
Wang et al. (2026) studied this question.