Application of DeepSeek-based AI teaching assistant in teaching anesthesiology theories

Medical education, particularly in specialized fields like anesthesiology, faces challenges of extensive knowledge systems and low teaching efficiency. Traditional teaching methods often fail to meet individualized learning needs and effectively assess students’ mastery of key learning points. Anesthesiology education is further complicated by its broad learning points and densely constructed knowledge systems, yet it occupies a small portion of undergraduate medical curricula. Recent advancements in generative artificial intelligence (AI) offer promising solutions to enhance medical education. This study explores the application of a DeepSeek-based AI teaching assistant in teaching anesthesiology theories. A single-center, randomized controlled trial was conducted with 48 participants, including 24 fourth-year medical students and 24 non-anesthesiology resident physicians. Participants were stratified by trainee type, then randomized 1:1 via a random-number table. Whereas the control group received traditional teaching (lectures, Q the experimental group received traditional teaching (same as the control group) plus 24/7 online support, personalized learning plans, and real-time performance analysis from the AI assistant. The experimental group achieved significantly higher theoretical test scores immediately after the course compared to the control group (P = 0.007). However, no significant difference was observed in knowledge retention one month later (P = 0.277). Student satisfaction surveys showed significantly higher scores in learning engagement and responsiveness to questions for the experimental group (P < 0.05), but no significant differences in learning interest or satisfaction. The DeepSeek-based AI teaching assistant enhanced short-term knowledge acquisition and learning engagement in anesthesiology education. However, long-term knowledge retention and overall satisfaction were not significantly improved. Future studies with larger sample sizes and longer observation periods are needed to further validate the effectiveness of AI teaching assistants in medical education and explore their potential in promoting clinical skill training.