Premium
SU‐E‐J‐252: A Motion Algorithm to Extract Physical and Motion Parameters of a Mobile Target in Cone‐Beam Computed Tomographic Imaging Retrospective to Image Reconstruction
Author(s) -
Ali I,
Alsbou N,
Ahmad S
Publication year - 2014
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4888306
Subject(s) - imaging phantom , amplitude , cone beam computed tomography , iterative reconstruction , medical imaging , tomography , motion (physics) , physics , image quality , computer vision , mathematics , algorithm , artificial intelligence , computer science , optics , image (mathematics) , computed tomography , medicine , radiology
Purpose: A motion algorithm was developed to extract actual length, CT‐numbers and motion amplitude of a mobile target imaged with cone‐beam‐CT (CBCT) retrospective to image‐reconstruction. Methods: The motion model considered a mobile target moving with a sinusoidal motion and employed three measurable parameters: apparent length, CT number level and gradient of a mobile target obtained from CBCT images to extract information about the actual length and CT number value of the stationary target and motion amplitude. The algorithm was verified experimentally with a mobile phantom setup that has three targets with different sizes manufactured from homogenous tissue‐equivalent gel material embedded into a thorax phantom. The phantom moved sinusoidal in one‐direction using eight amplitudes (0–20mm) and a frequency of 15‐cycles‐per‐minute. The model required imaging parameters such as slice thickness, imaging time. Results: This motion algorithm extracted three unknown parameters: length of the target, CT‐number‐level, motion amplitude for a mobile target retrospective to CBCT image reconstruction. The algorithm relates three unknown parameters to measurable apparent length, CT‐number‐level and gradient for well‐defined mobile targets obtained from CBCT images. The motion model agreed with measured apparent lengths which were dependent on actual length of the target and motion amplitude. The cumulative CT‐number for a mobile target was dependent on CT‐number‐level of the stationary target and motion amplitude. The gradient of the CT‐distribution of mobile target is dependent on the stationary CT‐number‐level, actual target length along the direction of motion, and motion amplitude. Motion frequency and phase did not affect the elongation and CT‐number distributions of mobile targets when imaging time included several motion cycles. Conclusion: The motion algorithm developed in this study has potential applications in diagnostic CT imaging and radiotherapy to extract actual length, size and CT‐numbers distorted by motion in CBCT imaging. The model provides further information about motion of the target.