Surgical endoscopy · 2026
Feasibility of using artificial intelligence models to identify the obturator nerve in laparoscopic selective lateral pelvic lymph node dissection.
СтатьяПоддержка исследования: вне правительства СШАНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Zeng Hao, Wang Xiaojie, Pan Hongfeng, Zhu Heyuan, Zhu Baihan, Sun Yanwu, Tang Yongqin, Chi Pan, Huang Ying
Аннотация
BACKGROUND: Protecting the obturator nerve (ON) is crucial during selective lateral pelvic lymph node dissection (SLPLND). This study evaluated the feasibility of an artificial intelligence (AI)-assisted system for real-time intraoperative identification of the ON in laparoscopic surgery.
METHODS: This study retrospectively analyzed surgical video from 30 patients who underwent laparoscopic SLPLND between June 2023 and December 2024. The obturator nerves in selective lateral pelvic lymph node dissection videos were analyzed using FDIM HoloSurg software. The panel of experts consisted of three senior surgeons with extensive experience in SLPLND, who analyzed the videos and scored them using a Likert scale (0, very poor; 1, poor; 2, acceptable; 3, good; and 4, excellent).
RESULTS: This research successfully established a real-time obturator nerve recognition model based on YOLOv11. The model was trained and validated using a dataset consisting of 1,530 high-quality surgical images, which were divided into training and validation sets at a ratio of 7:3. It achieved an overall accuracy of 0.962, a precision of 0.781, a recall of 0.775, and an F1 score of 0.778. Thirty challenging scenarios were selected from the validation cohort. In the vast majority of cases (73.3%), this model achieved an excellent score, while the scores obtained by junior surgeons were generally lower (P < 0.001).
CONCLUSION: This study confirms the technical feasibility of an AI-based real-time system for assisting in the identification of the ON. The system demonstrates a balance between high recall and precision, which can assist junior surgeons in identifying the ON and enhancing surgical safety.
METHODS: This study retrospectively analyzed surgical video from 30 patients who underwent laparoscopic SLPLND between June 2023 and December 2024. The obturator nerves in selective lateral pelvic lymph node dissection videos were analyzed using FDIM HoloSurg software. The panel of experts consisted of three senior surgeons with extensive experience in SLPLND, who analyzed the videos and scored them using a Likert scale (0, very poor; 1, poor; 2, acceptable; 3, good; and 4, excellent).
RESULTS: This research successfully established a real-time obturator nerve recognition model based on YOLOv11. The model was trained and validated using a dataset consisting of 1,530 high-quality surgical images, which were divided into training and validation sets at a ratio of 7:3. It achieved an overall accuracy of 0.962, a precision of 0.781, a recall of 0.775, and an F1 score of 0.778. Thirty challenging scenarios were selected from the validation cohort. In the vast majority of cases (73.3%), this model achieved an excellent score, while the scores obtained by junior surgeons were generally lower (P < 0.001).
CONCLUSION: This study confirms the technical feasibility of an AI-based real-time system for assisting in the identification of the ON. The system demonstrates a balance between high recall and precision, which can assist junior surgeons in identifying the ON and enhancing surgical safety.