Abstract：
OBJECTIVE: Transoral robotic surgery (TORS) is a challenging procedure due to its small workspace and complex anatomy. Ultrasound (US) image guidance has the potential to improve surgical outcomes, but an appropriate method for US probe manipulation has not been defined. This study evaluates using an additional robotic (4th) arm on the da Vinci Surgical System to perform extracorporeal US scanning for image guidance in TORS.
METHODS: A stereoscopic imaging system and da Vinci-compatible US probe attachment were developed to enable control of the extracorporeal US probe from the surgeon console. The prototype was compared to freehand US by nine operators in three tasks on a healthy volunteer: (1) identification of the common carotid artery, (2) carotid artery scanning, and (3) identification of the submandibular gland. Operator workload and user experience were evaluated using a questionnaire.
RESULTS: The robotic US tasks took longer than freehand US tasks (2.09x longer; p = 0.001 ) and had higher operator workload (2.12x higher; p = 0.004 ). However, operator-rated performance was closer (avg robotic/avg freehand = 0.66; p = 0.017 ), and scanning performance measured by MRI-US average Hausdorff distance provided no statistically significant difference.
CONCLUSIONS: Extracorporeal US scanning for intraoperative US image guidance is a convenient approach for providing the surgeon direct control over the US image plane during TORS, with little modification to the existing operating room workflow. Although more time-consuming and higher operator workload, several methods have been identified to address these limitations.
METHODS: A stereoscopic imaging system and da Vinci-compatible US probe attachment were developed to enable control of the extracorporeal US probe from the surgeon console. The prototype was compared to freehand US by nine operators in three tasks on a healthy volunteer: (1) identification of the common carotid artery, (2) carotid artery scanning, and (3) identification of the submandibular gland. Operator workload and user experience were evaluated using a questionnaire.
RESULTS: The robotic US tasks took longer than freehand US tasks (2.09x longer; p = 0.001 ) and had higher operator workload (2.12x higher; p = 0.004 ). However, operator-rated performance was closer (avg robotic/avg freehand = 0.66; p = 0.017 ), and scanning performance measured by MRI-US average Hausdorff distance provided no statistically significant difference.
CONCLUSIONS: Extracorporeal US scanning for intraoperative US image guidance is a convenient approach for providing the surgeon direct control over the US image plane during TORS, with little modification to the existing operating room workflow. Although more time-consuming and higher operator workload, several methods have been identified to address these limitations.
摘要：
目的：经口机器人手术（TORS）由于工作空间小，解剖结构复杂，是一项具有挑战性的程序。超声（US）图像引导具有改善手术效果的潜力，但美国探针操作的适当方法尚未定义。这项研究评估了在达芬奇手术系统上使用额外的机器人（第四）臂进行体外US扫描，以在TORS中进行图像指导。
方法：开发了一种立体成像系统和与达芬奇兼容的US探头附件，以使外科医生控制台能够控制体外US探头。九位操作员在健康志愿者的三项任务中将原型与徒手US进行了比较：（1）识别颈总动脉，(2)颈动脉扫描,(3)颌下腺的鉴定。使用问卷评估操作员工作量和用户体验。
结果：机器人US任务比徒手US任务花费的时间更长（2.09倍；p=0.001），并且操作员的工作量更高（高2.12倍；p=0.004）。然而,操作员额定性能更接近(avg机器人/avg徒手=0.66；p=0.017)，通过MRI-US平均Hausdorff距离测量的扫描性能没有统计学上的显着差异。
结论：用于术中US图像指导的体外US扫描是一种方便的方法，可在TORS期间为外科医生提供对US图像平面的直接控制，对现有的手术室工作流程几乎没有修改。虽然更耗时和更高的操作员工作量，已经确定了几种方法来解决这些限制。
方法：开发了一种立体成像系统和与达芬奇兼容的US探头附件，以使外科医生控制台能够控制体外US探头。九位操作员在健康志愿者的三项任务中将原型与徒手US进行了比较：（1）识别颈总动脉，(2)颈动脉扫描,(3)颌下腺的鉴定。使用问卷评估操作员工作量和用户体验。
结果：机器人US任务比徒手US任务花费的时间更长（2.09倍；p=0.001），并且操作员的工作量更高（高2.12倍；p=0.004）。然而,操作员额定性能更接近(avg机器人/avg徒手=0.66；p=0.017)，通过MRI-US平均Hausdorff距离测量的扫描性能没有统计学上的显着差异。
结论：用于术中US图像指导的体外US扫描是一种方便的方法，可在TORS期间为外科医生提供对US图像平面的直接控制，对现有的手术室工作流程几乎没有修改。虽然更耗时和更高的操作员工作量，已经确定了几种方法来解决这些限制。
