基于虚拟现实技术的机器人肾盂输尿管成形术分段式培训体系的建立和应用

目的：建立一种基于虚拟现实技术的分段式培训体系，并应用于机器人肾盂输尿管成形术（robotic-assisted laparoscopic pyeloplasty，RALP）的手术培训。 方法：首先采用虚拟现实技术，通过机器人模拟培训系统对3名泌尿外科医师进行基础操作培训和缝合强化培训，对比培训前后手术医师在吻合时间等关键指标的变化。其次将RALP分为游离、裁剪、缝合、置管等4个分段，受训医师通过机器人系统参与真实手术操作，先以分段为单位完成学习曲线，再独立完成完整的RALP手术。 结果：经虚拟现实培训后，受训医师的吻合时间（141.0 ± 9.0 s vs. 312.6 ± 34.5 s）和运动距离（140.9 ± 16.2 cm vs. 323.1 ± 50.6 cm）显著缩短（P<0.01），机械臂碰撞次数（1.6 ± 0.6 vs. 7.1 ± 1.0）和器械离开视野次数（0.8 ± 0.5 vs. 3.2 ± 0.9）显著减少（P<0.05），脱离目标次数在训练前后无明显变化。通过每人两轮的分段式真实手术训练后，受训医师能够独立完成RALP手术，手术时间要长于具备成熟RALP手术经验医师的同期手术时间（149 ± 36 min vs. 117 ± 10 min），但差异无统计学意义（P＝0.060）。两者在术中出血量、漏尿发生率和术后住院时间等围手术期指标上无显著差异。 结论：通过将基于虚拟现实技术的分段式培训体系应用于RALP的手术培训，可以在保证手术安全性和质量的前提下提高受训医师的操作技水平，具备独立完成RALP手术的能力。

Establishment and application of a segmented training system for robot-assisted laparoscopic pyeloplasty based on virtual reality technology

Objective: To establish a virtual reality technology-based segmented training system and apply it to the surgical training of robotic-assisted laparoscopic pyeloplasty (RALP). Methods: First, virtual reality technology was used to conduct basic operation training and intensive training of suture for 3 urologists through the robotic simulation training system. Second, the urologists in training participated in the real RALP operation through the robotic system. The surgery was divided into 4 sections including separating, cutting, suturing, and indwelling. Finally, the complete RALP operation would be independently conducted by trainees, following the completion of learning curve in segmented training. Results: After virtual reality training, the anastomotic time (141.0 ± 9.0 s vs. 312.6 ± 34.5 s) and motion distance (140.9 ± 16.2 cm vs. 323.1 ± 50.6 cm) were significantly shortened (P<0.01), and the times of both instrument collisions (1.6 ± 0.6 vs. 7.1 ± 1.0) and instrument out of sight (0.8 ± 0.5 vs. 3.2 ± 0.9) were significantly reduced (P<0.05). However, the times of missed target did not change significantly before and after training. After two rounds of segmented training per person, trainees were able to conduct RALP independently. Their operation time was longer than that of senior surgeons (149 ± 36 min vs. 117 ± 10 min), however, the difference was not statistically significant (P=0.060). In addition, there was no significant difference in perioperative parameters such as intraoperative blood loss, incidence of urine leakage, and postoperative hospital stay. Conclusion: By applying the virtual reality technology-based segmented training system to the surgical training of RALP, it can improve the operational skills of trainees on the premise of ensuring the safety and quality of the operation.