Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Aug 20, 2024
Date Accepted: Dec 27, 2024
An automated process for monitoring of amiodarone treatment: Development and evaluation
ABSTRACT
Background:
Amiodarone treatment requires repeated laboratory evaluations of thyroid and liver function due to potential side effects. Robotic process automation utilizes software robots to automate repetitive and routine tasks.
Objective:
This study aimed to develop such a robot using a diagnostic classification algorithm for amiodarone follow-up.
Methods:
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 next laboratory testing and management suggestions for a physician, serving as a human-in-the-loop responsible for decisions. The robot’s performance was studied alongside the existing real-world manual follow-up routine for amiodarone treatment.
Results:
Results:
Following iterative technical improvements, a robot prototype was validated against physician orders (n=390 paired orders). The robot recommended a mean (SD) follow-up interval of 4.5 (2.4) months, compared to 3.1 (1.4) months by physicians (P<.001). For normal laboratory findings, the robot recommended a six-month follow-up in 72.2% of cases, whereas physicians did so in only 9.9% of cases, favoring a 3–4-month follow-up (58.5%). All patients diagnosed with new side effects (n=12) were correctly detected by the robot, whereas only 8 by the physician.
Conclusions:
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 labour, decrease the frequency of laboratory testing, and improve the detection of side effects, thereby reducing costs and enhancing patient value.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.