Quantitative evaluation of the performance of different deformable image registration algorithms in helical, axial, and cone-beam CT images using a mobile phantom.

The goal of this project is to investigate quantitatively the performance of different deformable image registration algorithms (DIR) with helical (HCT), axial (ACT), and cone-beam CT (CBCT). The variations in the CT-number values and lengths of well-known targets moving with controlled motion were evaluated. Four DIR algorithms: Demons, Fast-Demons, Horn-Schunck and Lucas-Kanade were used to register intramodality CT images of a mobile phantom scanned with different imaging techniques. The phantom had three water-equivalent targets inserted in a low-density foam with different lengths (10-40 mm) and moved with adjustable motion amplitudes (0-20 mm) and frequencies (0-0.5 Hz). The variations in the CT-number level, volumes and shapes of these targets were measured from the spread-out of the CT-number distributions. In CBCT, most of the DIR algorithms were able to produce the actual lengths of the mobile targets; however, the CT-number values obtained from the DIR algorithms deviated from the actual CT-number of the targets. In HCT, the DIR algorithms were successful in deforming the images of the mobile targets to the images of the stationary targets producing the CT-number values and lengths of the targets for motion amplitudes <20 mm. Similarly in ACT, all DIR algorithms produced the actual CT-number values and lengths of the stationary targets for low-motion amplitudes <15 mm. The optical flow-based DIR algorithms such as the Horn-Schunck and Lucas-Kanade performed better than the Demons and Fast-Demons that are based on attraction forces particularly at large motion amplitudes. In conclusion, most of the DIR algorithms did not reproduce well the CT-number values and lengths of the targets in images that have artifacts induced by large motion amplitudes. The deviations in the CT-number values and variations in the volume of the mobile targets in the deformed CT images produced by the different DIR algorithms need to be considered carefully in the treatment planning for accurate dose calculation dose coverage of the tumor, and sparing of critical structures.

A motion algorithm to extract physical and motion parameters of mobile targets from cone-beam computed tomographic images.

PURPOSE: A motion algorithm has been developed to extract length, CT number level and motion amplitude of a mobile target from cone-beam CT (CBCT) images. MATERIALS AND METHODS: The algorithm uses three measurable parameters: Apparent length and blurred CT number distribution of a mobile target obtained from CBCT images to determine length, CT-number value of the stationary target, and motion amplitude. The predictions of this algorithm are tested with mobile targets having different well-known sizes that are made from tissue-equivalent gel which is inserted into a thorax phantom. The phantom moves sinusoidally in one-direction to simulate respiratory motion using eight amplitudes ranging 0-20 mm. RESULTS: Using this motion algorithm, three unknown parameters are extracted that include: Length of the target, CT number level, speed or motion amplitude for the mobile targets from CBCT images. The motion algorithm solves for the three unknown parameters using measured length, CT number level and gradient for a well-defined mobile target obtained from CBCT images. The motion model agrees with the measured lengths which are dependent on the target length and motion amplitude. The gradient of the CT number distribution of the mobile target is dependent on the stationary CT number level, the target length and motion amplitude. Motion frequency and phase do not affect the elongation and CT number distribution of the mobile target and could not be determined. CONCLUSION: A motion algorithm has been developed to extract three parameters that include length, CT number level and motion amplitude or speed of mobile targets directly from reconstructed CBCT images without prior knowledge of the stationary target parameters. This algorithm provides alternative to 4D-CBCT without requirement of motion tracking and sorting of the images into different breathing phases. The motion model developed here works well for tumors that have simple shapes, high contrast relative to surrounding tissues and move nearly in regular motion pattern that can be approximated with a simple sinusoidal function. This algorithm has potential applications in diagnostic CT imaging and radiotherapy in terms of motion management.