This work introduces a new method for verifying MLC leaf positions with enough spatial resolution to replace film-based methods in performing QA tests. It is implemented on a 2D ion chamber array, and it is based on the principle of varying signal response of a volumetric detector to partial irradiation. A PTW 2D-ARRAY seven29 (PTW-729 2D) array was used to assess a Siemens OPTIFOCUS MLC. Partial volume response curves for chambers in the array were obtained by irradiating them with the leaves of the MLC, progressively covering varying portions of the chambers correlated with the leaf positions. The readings from the array's chambers are processed with an in-house program; it generates a reference response that translates readings into leaf positions. This principle allows discriminating errors in pairs of opposing leaves that could combine to cancel their detection with other tools.
Patterns of leaf positions, similar to the Bayouth test but with different, purposefully introduced errors, were generated and used to test the effectiveness of the method. The same patterns were exposed on radiographic film and analyzed with the RIT software for validation. For four test patterns with a total of 100 errors of Symbol, Symbol and Symbol, all were correctly determined with the proposed method. The analysis of the same pattern with film using the Bayouth routine in the RIT software resulted in either somewhat low true positives combined with a large fraction of false positives, or a low true positive rate with a low false positive ratio, the results being significantly affected by the threshold selected for the analysis.
This method provides an effective, easy to use tool for quantitative MLC QA assessment, with excellent spatial resolution. It can be easily applied to other 2D arrays, as long as they exhibit a partial volume detector response.
PACS number: 87.55.Qr