Computerized tomography-based quality assurance tool for proton range compensators

Abstract

The authors have developed an automatic quality assurance method based on computerized tomography (CT) data for physically manufactured proton range compensators. Eight range compensators used for proton therapy at their institution were scanned using CT and their depth distributions were compared with data provided by a proton treatment planning system (TPS). The depth difference (DD), distance to agreement (DTA), and a composite analysis (CA) indicating regions that passed either of the tests were used for depth verification. When the tolerance limits were set at 3 mm for DD and 1 mm for DTA, the average percentages of points exceeding the acceptance criteria for DD, DTA, and CA for the eight compensators were 9.0%, 5.3% and 3.2%, respectively. In general, the percentages of points exceeding the acceptance criteria were much higher in DD than in DTA analysis and mismatch of depth was greater in regions with high depth gradients. This is considered to arise mainly because of systematic errors such as the physical size of the drill, slight bending or slipping of the drill bit near high gradient regions, the limit of CT resolution, and the uncertainty in CT number. When the tolerance limit of DTA was increased from 1 to 2 mm, the average percentage of points exceeding the acceptance criteria for DTA was reduced from 5.3% to 0.9%. Their results show that depth comparison of proton range compensators based on CT images can provide more systematic data than does the current method, in which a representative group of drill points is measured manually.

Is there a clinical benefit with a smooth compensator design compared with a plunged compensator design for passive scattered protons?

“…One limitation in the TPS's design of the compensator is the inability to accurately model the dimensions of the drill bit. 4,5,8 The tapering of the drill bit is not taken into account for the plunge technique, which results in underestimation the material removed during the milling process. This could result in discrepancies between the calculated and measured dose.…”

“…7 It is also important to note that the drill bit tapering is not modeled by the TPS compensator design algorithm, which can increase uncertainties, especially in sharp-depth gradient regions. 4,8 This error does not cause a large dose deviation in regions of small thickness change, but it is a known source of systematic error. 5 The tapering effect is more pronounced for deeper compensators because more material is removed than accounted for in the compensator design pattern.…”

“…Ultimately, lack of a true drill bit model could result in calculated and measured dose discrepancies. 4,5,8 Alternatively, a "smooth" compensator design is available in the TPS. The plunged-depth points must be converted into a 3-dimensional wireframe surface so that the compensator can be milled with a smooth surface.…”

Development of a 3D optical scanning‐based automatic quality assurance system for proton range compensators

“…Several errors can occur during the CMM manufacturing process. 4 In particular, depth differences (DDs) caused by limitations in the size or tapering (angle) of the drill bit can extend over the entire RC, 6,7 and these errors must be fully characterized by precision QA.…”

“…5 However, this technique required extensive image processing to minimize random, line pattern, and background noise as well as blurring of the image due to x-ray scattering. Yoon et al 6 and Li et al, 7 have proposed computed tomography (CT)-based RC QA techniques, which require the use of a marker or the application of a threshold value on the CT images to determine a reference thickness. As a result, part of the RC geometry could be lost owing to the finite slice thickness of the CT scan.…”