Implementation and Evaluation of an Offline RPA-Based Scheduling Visualization Tool for Radiotherapy Under Security Constraints.

Abstract

Timely access to radiotherapy appointment information remains challenging in hospitals-including many in Japan-due to fragmented information systems and strict security policies restricting access to databases or application programming interfaces (API). We developed an offline robotic process automation (RPA) workflow that captures scheduling data directly from the graphical user interfaces (GUIs) of the electronic medical record (EMR) and radiology information system (RIS) consolidating new-patient consultations, computed tomography (CT) simulations, and first-fraction irradiation sessions into a color-coded, four-week Excel calendar. The robot, scripted using the no-code KEYENCE RK-10 environment, logs into each application, applies rule-based filters, exports comma-separated-values (CSV) files, and populates a pre-formatted template without requiring any back-end modifications. During a two-week deployment at a community radiotherapy center, each weekday run finished in a median 3.9 min [inter-quartile range, IQR 3.1-4.8 min] and listed 36 appointments (16 consultations, five CTs, and 15 first-fraction irradiations). Two discrepancies compared to a legacy whiteboard revealed manual omissions. A semi-structured survey of 12 staff members recorded the highest Likert score [median 5.0, IQR 4.0-5.0] for reductions in documentation time, perceived workload, and error frequency in favor of the RPA implementation. This lightweight GUI-level approach provides a secure and rapidly deployable solution for small- to medium-sized radiotherapy centers operating under stringent information-technology (IT) policies, enhancing scheduling accuracy and staff efficiency without database integration. EMR, RPA, Radiation Therapy, Automation, Visualization.