Johansson Birgitta I, Landahl Jonas, Tammelin Karin, Aerts Erik, Lundberg Christina E, Adiels Martin, Lindgren Martin, Rosengren Annika, Papachrysos Nikolaos, Filipsson Nyström Helena, Sjöland Helen
Department of Molecular and Clinical Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.
Department of Medicine, Geriatrics and Emergency Medicine, Sahlgrenska University Hospital, Gothenburg, Sweden.
J Med Internet Res. 2025 Feb 19;27:e65473. doi: 10.2196/65473.
Amiodarone treatment requires repeated laboratory evaluations of thyroid and liver function due to potential side effects. Robotic process automation uses software robots to automate repetitive and routine tasks, and their use may be extended to clinical settings.
Thus, this study aimed to develop a robot using a diagnostic classification algorithm to automate repetitive laboratory evaluations for amiodarone follow-up.
We designed a robot and clinical decision support system based on expert clinical advice and current best practices in thyroid and liver disease management. The robot provided recommendations on the time interval to follow-up laboratory testing and management suggestions, while the final decision rested with a physician, acting as a human-in-the-loop. The performance of the robot was compared to the existing real-world manual follow-up routine for amiodarone treatment.
Following iterative technical improvements, a robot prototype was validated against physician orders (n=390 paired orders). The robot recommended a mean follow-up time interval of 4.5 (SD 2.4) months compared to the 3.1 (SD 1.4) months ordered by physicians (P<.001). For normal laboratory values, the robot recommended a 6-month follow-up in 281 (72.1%) of cases, whereas physicians did so in only 38 (9.7%) of cases, favoring a 3- to 4-month follow-up (n=227, 58.2%). All patients diagnosed with new side effects (n=12) were correctly detected by the robot, whereas only 8 were by the physician.
An automated process, using a software robot and a diagnostic classification algorithm, is a technically and medically reliable alternative for amiodarone follow-up. It may reduce manual labor, decrease the frequency of laboratory testing, and improve the detection of side effects, thereby reducing costs and enhancing patient value.
由于潜在的副作用，胺碘酮治疗需要对甲状腺和肝功能进行反复的实验室评估。机器人流程自动化使用软件机器人来自动执行重复性和常规任务，其应用可能会扩展到临床环境。
因此，本研究旨在开发一种使用诊断分类算法的机器人，以自动执行胺碘酮随访的重复性实验室评估。
我们根据甲状腺和肝病管理方面的专家临床建议和当前最佳实践设计了一个机器人和临床决策支持系统。该机器人提供了实验室检测随访时间间隔的建议和管理建议，而最终决策由医生做出，医生作为“人工干预环节”。将该机器人的性能与胺碘酮治疗现有的实际手动随访常规进行了比较。
经过反复的技术改进，一个机器人原型针对医生的医嘱（n = 390组配对医嘱）进行了验证。机器人建议的平均随访时间间隔为4.5（标准差2.4）个月，而医生开出的时间间隔为3.1（标准差1.4）个月（P <.001）。对于正常的实验室值，机器人在281例（72.1%）病例中建议6个月随访，而医生仅在38例（9.7%）病例中这样做，更倾向于3至4个月随访（n = 227，58.2%）。机器人正确检测出所有被诊断为新副作用的患者（n = 12），而医生仅检测出8例。
使用软件机器人和诊断分类算法的自动化流程是胺碘酮随访在技术和医学上可靠的替代方案。它可以减少人工劳动，降低实验室检测频率，并提高副作用的检测率，从而降低成本并提高患者价值。
