High spatial resolution reconstruction technique for spect using a fan-beam collimator

T. Ichihara, K. Nambu, N. Motomura

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)

Abstract

The physical characteristics of the collimator cause degradation of resolution with increasing distance from the collimator surface. A new convolutional backprojection algorithm has been derived for fan beam SPECT data without rebinding into parallel beam geometry. The projections are filtered and then backprojected into the area within an isosceles triangle whose vertex is the focal point of the fan-beam and whose base is the fan-beam collimator face, and outside of the circle whose center is located midway between the focal point and the center of rotation and whose diameter is the distance between the focal point and the center of rotation. Consequently the backprojected area is close to the collimator surface. This algorithm has been implemented on a GCA-9300A SPECT system showing good results with both phantom and patient studies. The SPECT transaxial resolution was 4.6mm FWHM (reconstructed image matrix size of 256x256) at the center of SPECT FOV using UHR (ultra-high-resolution) fan beam collimators for brain study. Clinically, Tc-99m HMPAO and Tc-99m ECD brain data were reconstructed using this algorithm. The reconstruction results were compared with MRI images of the same slice position and showed significantly improved over results obtained with standard reconstruction algorithms.

Original languageEnglish
Pages (from-to)1149-1157
Number of pages9
JournalIEEE Transactions on Nuclear Science
Volume40
Issue number4
DOIs
Publication statusPublished - 08-1993

All Science Journal Classification (ASJC) codes

  • Nuclear and High Energy Physics
  • Nuclear Energy and Engineering
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'High spatial resolution reconstruction technique for spect using a fan-beam collimator'. Together they form a unique fingerprint.

Cite this