Liu Zhanhe, Petersen Lydia, Zhang Ziyang, Singapogu Ravikiran
Annu Int Conf IEEE Eng Med Biol Soc. 2020 Jul;2020:6090-6094. doi: 10.1109/EMBC44109.2020.9176158.
Cannulation is a routine yet challenging medical procedure resulting in a direct impact on patient outcomes. While current training programs provide guidelines to learn this complex procedure, the lack of objective and quantitative feedback impedes learning this skill more effectively. In this paper, we present a simulator for performing hemodialysis cannulation that captures the process using multiple sensing modalities that provide a multi-faceted assessment of cannulation. Further, we describe an algorithm towards segmenting the cannulation process using specific events in the sensor data for detailed analysis. Results from three participants with varying levels of clinical cannulation expertise are presented along with a metric that successfully differentiates the three participants. This work could lead to sensor-based cannulation skill assessment and training in the future potentially resulting in improved patient outcomes.
插管是一种常规但具有挑战性的医疗程序，会对患者的治疗结果产生直接影响。虽然当前的培训计划提供了学习这一复杂程序的指导方针，但缺乏客观和定量的反馈阻碍了更有效地学习这项技能。在本文中，我们展示了一种用于进行血液透析插管的模拟器，该模拟器使用多种传感模式来捕捉过程，这些模式对插管提供多方面的评估。此外，我们描述了一种算法，该算法利用传感器数据中的特定事件对插管过程进行分割，以便进行详细分析。展示了三名具有不同临床插管专业水平的参与者的结果，以及一个成功区分这三名参与者的指标。这项工作可能会在未来带来基于传感器的插管技能评估和培训，从而有可能改善患者的治疗结果。
