Title :
Contrast enhancement and artifact reduction for projected index computed tomography
Author :
Zysk, Adam M. ; Marks, Daniel L. ; Carney, P. Scott ; Boppart, Stephen A.
Author_Institution :
Dept. of Electr. & Comput. Eng., Illinois Univ., Champaign, IL, USA
Abstract :
The projected index computed tomography (PICT) imaging modality provides a computationally reconstructed image of optical index of refraction that is free from spatial distortions. Current implementation using standard filtered backprojection techniques contain linear artifacts resulting from beam path refraction at boundaries with a significant index change. We present the implementation of a least-squares solution whereby every projection datum is assigned a confidence weight, affecting the magnitude of its inclusion into the reconstructed image. The algorithm is shown to significantly reduce artifacts, hence increasing system sensitivity while maintaining spatial accuracy.
Keywords :
biomedical optical imaging; computerised tomography; image enhancement; image reconstruction; least squares approximations; medical image processing; refractive index; beam path refraction; computationally reconstructed image; confidence weight; contrast enhancement; imaging system sensitivity; least-squares solution; linear artifact reduction; optical refractive index; projected index computed tomography imaging; spatial accuracy; spatial distortions; standard filtered backprojection techniques; Computed tomography;
Conference_Titel :
Biomedical Imaging: Nano to Macro, 2004. IEEE International Symposium on
Print_ISBN :
0-7803-8388-5
DOI :
10.1109/ISBI.2004.1398734