An Interface for Analysis of Medical Linear Accelerator Parameters Using Process Behavior Charts

Abstract

Modern medical linear accelerators are complex digital devices capable of producing photon and electron beams of multiple energies. These accelerators are comprised of a number of subsystems that facilitate the treatment of cancer using an expanding range of clinical approaches. Accelerator design has interwoven hardware and software such that the opportunity to control components and systems with increasing precision has also increased the ability to monitor system operation. While the manufacturing and design of medical electron accelerators has improved the reliability and consistency of operation, system dysfunction and failure still occur. Currently, digital data accumulated in the accelerator system are not being exploited in a systematic manner. Linear accelerator interlocks ensure that the operation of the system is discontinued when parameters exceed the limits of a system's operating specifications. Component failure or dysfunction requires immediate repair and service engineering on site. The result is unscheduled machine downtime and disruption of clinical services. The purpose of this thesis was to develop an effective process for detecting unexpected deviations in accelerator system operating parameters and/or performance that predicts component failure or system dysfunction. The focus of this thesis has three major objectives: (1) Assist in the design of the Statistical Process Control (SPC) database of linear accelerator parameters to detect/predict performance dysfunction; (2) Develop an interface to efficiently display, manage and report the SPC results of all parameters stored in the database; (3) Determine the effectiveness of the standard Individual/Moving Range (I/MR) tests to detect performance dysfunction prior to the actuation of interlocks for several of the parameters under surveillance and the consistency of these parameters being monitored.