Share Email Print

Proceedings Paper

Limited-angle reconstruction algorithms in computed-tomographic microscopic imaging
Author(s): Ravil Chamgoulov; Michael Tsiroulnikov; Pierre Lane; Calum MacAulay
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

We present our results on development of algorithms for image reconstruction in optical computed-tomography microscopy. The optical computed-tomography microscope with light modulation, recently developed by our group at the BC Cancer Research Centre is a novel imaging device for three-dimensional visualization and quantitative analysis of absorption-stained biological samples. The angles for projection in the system are limited to the range of 0 ≤ φ ≤ 135° by the numerical aperture of the illumination objective. For the limited-angle tomography problem we have developed several reconstruction algorithms. One algorithm is based on the Radon transformation and assumes parallel ray projections. In order to compensate for limited data, several reconstructions are generated from several sets of projections of a specimen, acquired at different orientations of parallel-ray light scanning. The reconstructions are combined together using a vote criteria to create a final volume. Another reconstruction algorithm developed by the group employs both transform-based and iterative methods to address the limited-angle reconstruction problem. In this algorithm the transform-based method is used as an initial starting point for the following iterative reconstruction. A feedback correction of the reconstruction image is made on each iteration step. The method enables to incorporate previously known information about the object into the reconstruction process. The algorithm improves reconstruction accuracy at a reasonable computational cost and programming commitment. Three-dimensional microscopic images of quantitatively absorption-stained cells have been reconstructed with the resolution better than 6 microns.

Paper Details

Date Published: 29 April 2005
PDF: 8 pages
Proc. SPIE 5747, Medical Imaging 2005: Image Processing, (29 April 2005); doi: 10.1117/12.594708
Show Author Affiliations
Ravil Chamgoulov, BC Cancer Research Ctr. (Canada)
Michael Tsiroulnikov, MIKET DSP Solutions (Canada)
Pierre Lane, BC Cancer Research Ctr. (Canada)
Calum MacAulay, BC Cancer Research Ctr. (Canada)

Published in SPIE Proceedings Vol. 5747:
Medical Imaging 2005: Image Processing
J. Michael Fitzpatrick; Joseph M. Reinhardt, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?