Abstract
The quality of adaptive treatment planning depends on the accuracy of its underlying deformable image registration (DIR). The purpose of this study is to evaluate the performance of two DIR algorithms, B‐spline‐based deformable multipass (DMP) and deformable demons (Demons), implemented in a commercial software package. Evaluations were conducted using both computational and physical deformable phantoms. Based on a finite element method (FEM), a total of 11 computational models were developed from a set of CT images acquired from four lung and one prostate cancer patients. FEM generated displacement vector fields (DVF) were used to construct the lung and prostate image phantoms. Based on a fast‐Fourier transform technique, image noise power spectrum was incorporated into the prostate image phantoms to create simulated CBCT images. The FEM‐DVF served as a gold standard for verification of the two registration algorithms performed on these phantoms. The registration algorithms were also evaluated at the homologous points quantified in the CT images of a physical lung phantom. The results indicated that the mean errors of the DMP algorithm were in the range of 1.0~3.1mm for the computational phantoms and 1.9 mm for the physical lung phantom. For the computational prostate phantoms, the corresponding mean error was 1.0–1.9 mm in the prostate, 1.9–2.4 mm in the rectum, and 1.8–2.1 mm over the entire patient body. Sinusoidal errors induced by B‐spline interpolations were observed in all the displacement profiles of the DMP registrations. Regions of large displacements were observed to have more registration errors. Patient‐specific FEM models have been developed to evaluate the DIR algorithms implemented in the commercial software package. It has been found that the accuracy of these algorithms is patient‐dependent and related to various factors including tissue deformation magnitudes and image intensity gradients across the regions of interest. This may suggest that DIR algorithms need to be verified for each registration instance when implementing adaptive radiation therapy.PACS numbers: 87.10.Kn, 87.55.km, 87.55.Qr, 87.57.nj
Highlights
178 Stanley et al.: Evaluation of deformable image registration algorithms changes may compromise the accuracy of dose calculation for each organ
Depending on the registration techniques used, the transformation map can be represented by different mathematical models such as affine transform,(5) thin-plate spline,(6) or B-spline basis,(7) or adapted directly through optical flow-based equations.[8]. Similarity metrics can be represented in different forms including the sum of squared difference, cross-correlation, or normalized mutual information.[9,10] Like dose calculation algorithms, these deformable image registration (DIR) algorithms must be thoroughly evaluated before they are used in clinic for adaptive radiation therapy
As reported by Kashani et al[18] and Liu et al,(19) large registration errors can be observed in regions of uniform image intensity, and the above evaluations are limited by the number of the objects being tracked; so errors estimated by the feature-guided evaluation methods may not be representative of the registration accuracy in voxels at a distance from those landmarks.[20]
Summary
178 Stanley et al.: Evaluation of deformable image registration algorithms changes may compromise the accuracy of dose calculation for each organ. Adaptive radiotherapy (ART) aims to minimize the dosimetric impact of anatomical changes by reoptimizing the original treatment plan if its quality degrades.[3] A key step in the implementation of ART is to match each point on daily CT images to their correspondent points in the planning image This process can be accomplished with deformable image registration (DIR) techniques.[4] DIR is to derive a transformation map by maximizing the intensity similarity between the two images being registered. Castillo et al[17] developed an automatic method to identify and track landmark points in lung patient datasets These studies provided quantitative evaluation results on the performances of different DIR algorithms at these distinctive landmarks or their nearby regions. As reported by Kashani et al[18] and Liu et al,(19) large registration errors can be observed in regions of uniform image intensity, and the above evaluations are limited by the number of the objects being tracked; so errors estimated by the feature-guided evaluation methods may not be representative of the registration accuracy in voxels at a distance from those landmarks.[20]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
