Author: William N. Duggar, Li Yuan 👨🔬
Affiliation: University of Mississippi Medical Center 🌍
Purpose:
Radiation Oncology departments typically utilize various systems from different vendors. Ensuring the integrity and correctness of data during transfers between these systems is essential for safe and accurate treatment delivery. In alignment with AAPM TG-100 recommendations for clinical practices, a feasible, extendable, and low-cost web-based application was developed to minimize the second-check workload of medical physicists and reduce the need for human intervention.
Methods:
The project was initiated in response to a request from the medical physics team to create a system that improves the efficiency of objective second-check tasks. The facility’s IT department set up a web server with access to all relevant database servers. A simple web-based graphical user interface (GUI) was developed to accommodate different user roles.
Results:
Connections to the RayStation and MOSAIQ databases were successfully established using server names, usernames, passwords, and database names. SQL scripts for comparing patient names, treatment plans, prescriptions, and detailed field information were created by the database administrator based on physicists' requirements.
Physicists can now query and compare patient names, plan details, prescriptions, and field information from RayStation and MOSAIQ simply by entering the patient's MRN. Results are returned within three seconds, significantly reducing the time required for this task, which previously took an experienced physicist approximately three minutes to complete manually. Based on the AAPM 2023 salary survey (average primary salary: $229,700), a typical clinic (50 new patients/week) can save approximately $1,100 per month. However, this is just the very beginning phase of this system.
Conclusion:
This web-based application stands to significantly enhance the efficiency and capacity of medical physicists' second-check workflows without requiring substantial time or budget investments. Additionally, the system is designed to be expanded to handle tasks beyond human capability, even potentially leveraging machine learning algorithms to unlock its full potential.