Real-time non-intrusive workload monitoring-Integration of human factors in surgery training and assessment

项目摘要/摘要(30行) 在去耦合的外科工作需求中需要较高的生理和认知工作负荷 对患者预后、手术疗效和手术效果有显著影响。作为一种新的外科手术 技术，例如远程外科手术，得到了发展，外科手术将变得更加复杂，精神上 而且对外科医生的实际需求可能会增加，这使得开发远程和联网至关重要 安全有效的外科手术程序设计、测试和培训的工作量监测方法。这 Work将实施新技术和机器学习分析，以量化实时和远程 工作量和测试工作量反馈如何影响远程手术和外科手术中的护理提供 模拟环境。我们的总体假设是远程外科中的连接传感技术 程序和模拟可以提高外科培训和对其工作量影响的了解 性能；最终改善患者健康、手术效率和患者访问(例如，远程指导) 外科护理。提出了两个具体的目标来研究这一假说。 具体目标1目标是开发一种连接的传感器系统以客观地量化 模拟远程机器人程序中的工作负载实时。这包括：1)将非侵入式传感器集成到 模拟训练器或环境中的单个系统，2)训练机器学习技术以 使用模拟手术技能任务客观区分工作负载，以及3)验证各种不同的指标 医学实习生和专家参与者在不同任务难度下的认知负荷水平。 具体目标2的目标是确定实时工作负载反馈的影响 对实习生表现时间、错误和术中工作量的干预。提出了两项任务：1) 探索外科医生首选的提供工作量实时反馈的方法，并2)评估影响 关于任务表现和学习的工作量反馈。我们的主要假设是表演时间和 向参与者提供关于工作负载与性能的实时反馈时，错误会有所改善 没有任何反馈。 预期交付内容包括1)工作负载监控技术、算法和软件，用于 补充当前基于模拟的培训，2)目标和自动化工作负荷指标，3)实时 辅助干预工具，以及4)工作量监测对培训影响的初步证据。这个 这项拟议工作中的技术将通过减少人为因素造成的不良事件来改善公共健康 通过可以自适应地培训外科医生的干预技术，改善获得外科护理的机会 并远程评估熟练程度。

Project Summary/Abstract (30 lines) High physiological and cognitive workload required in de-coupled surgical work demands may have significant impact on patient outcome, surgical efficacy, and surgical performance. As novel surgical techniques, e.g., telesurgery, are developed, surgical operations will become more complex and the mental and physical demand on surgeons will likely increase, making it critical to develop remote and connected workload monitoring methods for the safe and effective surgical procedure design, testing, and training. This work will implement novel technology and machine learning analytics to quantify real-time and remote workload and test how workload feedback can impact care delivery in both in telesurgery and surgical simulation environments. Our overall hypothesis is that connected sensing technology in telesurgical procedures and simulation can improve surgical training and understanding of the impact of their workload on performance; ultimately improving patient health, surgery efficacy, and patient access (e.g., tele-mentoring) to surgical care. Two specific aims are proposed to investigate this hypothesis. The objective of Specific Aim 1 is to develop a connected sensor system to objectively quantify workload real-time in simulated telerobotic procedures. This involves: 1) integrating non-intrusive sensors into a single system within the simulation trainer or environment, 2) training machine learning techniques to objectively distinguish workload using a simulated surgical skills tasks, and 3) validating metrics across varying levels of cognitive loads under various task difficulty with medical trainees and expert participants. The objective of Specific Aim 2 is to determine the impact of the real-time workload feedback intervention on trainee performance times, errors, and intraoperative workload. Two tasks are proposed: 1) Explore modalities preferred by surgeons for providing real-time feedback on workload and 2) Assess impact of workload feedback on task performance and learning. Our primary hypothesis is that performance times and errors will improve when participants are provided realtime feedback on workload compared to performance with no feedback. The expected deliverables include 1) workload monitoring technology, algorithms, and software for complementing current simulation-based training, 2) objective and automated workload metrics, 3) real-time assistive intervention tool, and 4) preliminary evidence on impact of workload monitoring on training. The technology in this proposed work will improve public health by reducing adverse events due to human factors in surgery and improve access to surgical care with intervention technology that can adaptively train surgeons and remotely assess proficiency.