Title :
On geometric distortions in CT images
Author :
W. Zylka;H.-A. Wischmann
Author_Institution :
Div. of Tech. Syst., Philips Res. Lab., Hamburg, Germany
Abstract :
Based on the analysis of possible mechanical misalignments of a CT (computer tomography) unit, CT image distortions are modeled. By extending the model with respect to the individual image acquisition process, the influence of patient´s weight on CT table velocity is included as well as gentry and table motion tilts. The effect of non-linear bending of the CT table due to patient´s weight is investigated in detail. A method to determine the magnitude of geometric image distortions is proposed. It can be interpreted as an estimation of parameters of a non-linear model in three dimensional image space. The validity and applicability of the correction method, e.g. for ´frameless´ image-guided neurosurgery, is discussed. Applying the method to CT images of different patients, image distortions up to 2 mm have been detected.
Keywords :
"Computed tomography","Nonlinear distortion","Neurosurgery","Neoplasms","Engineering in Medicine and Biology Society","Solid modeling","Image enhancement","Electronic mail","Image analysis","Biology computing"
Conference_Titel :
Engineering in Medicine and Biology Society, 1996. Bridging Disciplines for Biomedicine. Proceedings of the 18th Annual International Conference of the IEEE
Print_ISBN :
0-7803-3811-1
DOI :
10.1109/IEMBS.1996.652736