Annals of gastroenterological surgery · 2025
Educational Impact of Artificial Intelligence-Navigation Surgery on Anatomical Landmark Recognition in Medical Students.
СтатьяНазвание 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.
Ninomiya Shigeo, Matsunobu Yusuke, Oyama Tabito, Takayama Hiroomi, Masuda Takashi, Hirashita Teijiro, Endo Yuichi, Tokuyasu Tatsushi, Inomata Masafumi
Аннотация
BACKGROUND: We evaluated the educational impact of artificial intelligence (AI)-navigation surgery for medical students which provides real-time anatomical landmark recognition during laparoscopic cholecystectomy (LC).
METHODS: Thirty fifth-year medical students were randomly assigned to three groups: surgeon-guided (n = 10), self-learning (n = 10), and AI-learning (n = 10). Each group annotated anatomical landmarks, extrahepatic bile duct (EHBD), cystic duct (CD), Rouvière's sulcus (RS), the base of liver segment 4 (S4), before and after training. The AI-learning group received real-time feedback using a deep learning segmentation model (HyperSeg). Learning outcomes were quantitatively assessed and compared to expert annotations using Dice coefficients, and post-study questionnaires were analyzed to evaluate understanding of anatomy and surgical procedures.
RESULTS: The mean Dice coefficients in the surgeon-guided (0.450 ± 0.025) and AI-learning groups (0.432 ± 0.038) were significantly higher in comparison to the self-learning group (0.351 ± 0.057, p = 0.00006). In an itemized analysis, significant improvements were observed in EHBD and RS recognition, but not in CD or S4 recognition. In the post-study questionnaire assessing anatomical understanding and the ability to comprehend the surgeon's perspective and intentions, the surgeon-guided group showed significantly better results in comparison to the self-learning group (p < 0.001 for each comparison). However, there was no significant difference between the AI-learning and self-learning groups.
CONCLUSIONS: AI has the potential to complement surgeon's guidance, reducing faculty burden while maintaining educational quality in surgical education.
METHODS: Thirty fifth-year medical students were randomly assigned to three groups: surgeon-guided (n = 10), self-learning (n = 10), and AI-learning (n = 10). Each group annotated anatomical landmarks, extrahepatic bile duct (EHBD), cystic duct (CD), Rouvière's sulcus (RS), the base of liver segment 4 (S4), before and after training. The AI-learning group received real-time feedback using a deep learning segmentation model (HyperSeg). Learning outcomes were quantitatively assessed and compared to expert annotations using Dice coefficients, and post-study questionnaires were analyzed to evaluate understanding of anatomy and surgical procedures.
RESULTS: The mean Dice coefficients in the surgeon-guided (0.450 ± 0.025) and AI-learning groups (0.432 ± 0.038) were significantly higher in comparison to the self-learning group (0.351 ± 0.057, p = 0.00006). In an itemized analysis, significant improvements were observed in EHBD and RS recognition, but not in CD or S4 recognition. In the post-study questionnaire assessing anatomical understanding and the ability to comprehend the surgeon's perspective and intentions, the surgeon-guided group showed significantly better results in comparison to the self-learning group (p < 0.001 for each comparison). However, there was no significant difference between the AI-learning and self-learning groups.
CONCLUSIONS: AI has the potential to complement surgeon's guidance, reducing faculty burden while maintaining educational quality in surgical education.