In a multivendor solution, the protocols on how systems interact are, by necessity, well dened and published. e design of a quality assurance (QA) program to monitor the correct hando of information from system to system is relatively straightforward; failure modes and their eects can be determined based on this information. e situation is more challenging for combined systems. e interaction between system components typically is handled internally via the vendor-provided soware. Computer-controlled delivery systems, although pre - venting manual entry errors, can also contribute to errors. e report of AAPM TG-35 (Purdy et al. 1993) discusses many aspects of QA of computer-controlled radiation delivery systems. e recent paper of Fraass (2008) makes a clear statement on the status of QA management at this time in our eld, and deserves to be cited in full: e knowledge, documentation, and understanding of control system architecture, design, and implementation that a typical user has is quite limited, making eective testing dicult. TG-35 recommends that vendors provide reasons for changes, bug x descriptions, modication details, operational changes, site-dependent and user-accessible data or soware which may be aected, testing procedures, revised specications, support documentation, operations manuals, and beta test results with any soware installation or update. Unfortunately, little of this is typically available, although vendor adherence to TG-35 recommendations would make possible signicantly more eective testing and use by the user.
ASJC Scopus subject areas
- Biochemistry, Genetics and Molecular Biology(all)
- Physics and Astronomy(all)