Application of a linear interpolation algorithm in radiation therapy dosimetry for 3D dose point acquisition

Abstract

AbstractAir-vented ion chambers are generally used in radiation therapy dosimetry to determine the absorbed radiation dose with superior precision. However, in ion chamber detector arrays, the number of array elements and their spacing do not provide sufficient spatial sampling, which can be overcome by interpolating measured data. Herein, we investigated the potential principle of the linear interpolation algorithm in volumetric dose reconstruction based on computed tomography images in the volumetric modulated arc therapy (VMAT) technique and evaluated how the ion chamber spacing and anatomical mass density affect the accuracy of interpolating new data points. Plane measurement doses on 83 VMAT treatment plans at different anatomical sites were acquired using Octavius 729, Octavius1500, and MatriXX ion chamber detector arrays, followed by the linear interpolation to reconstruct volumetric doses. Dosimetric differences in planning target volumes (PTVs) and organs at risk (OARs) between treatment planning system and reconstruction were evaluated by dose volume histogram metrics. The average percentage dose deviations in the mean dose (Dmean) of PTVs reconstructed by 729 and 1500 arrays ranged from 4.7 to 7.3% and from 1.5 to 2.3%, while the maximum dose (Dmax) counterparts ranged from 2.3 to 5.5% and from 1.6 to 7.6%, respectively. The average percentage dose/volume deviations of mixed PTVs and OARs in the abdomen/gastric and pelvic sites were 7.6%, 3.5%, and 7.2%, while mediastinum and lung plans showed slightly larger values of 8.7%, 5.1%, and 8.9% for 729, 1500, and MatriXX detector arrays, respectively. Our findings indicated that the smaller the spacing between neighbouring detectors and the more ion chambers present, the smaller the error in interpolating new data points. Anatomical regions with small local mass density inhomogeneity were associated with superior dose reconstruction. Given a large mass density difference in the various human anatomical structures and the characteristics of the linear interpolation algorithm, we suggest that an alternative data interpolation method should be used in radiotherapy dosimetry.