Spatial and temporal registration of CT and SPECT images: development and validation of a technique for in vivo three-dimensional semiquantitative analysis of bone

J Nucl Med. 2000 Jun;41(6):1075-81.

Abstract

The combined use of postoperative 3-dimensional CT and SPECT imaging provides a means of relating anatomy and physiology for the semiquantitative in vivo analysis of bone. This study focuses on the development and validation of a technique that accomplishes this through the registration of SPECT data to a 3-dimensional volume of interest (VOI) interactively defined on CT images.

Methods: Five human cadaver heads served as anthropomorphic models for all experiments. Four cranial defects were created in each specimen with inlay and onlay split-skull bone grafts reconstructed to skull and malar recipient sites. To acquire all images, each specimen was landmarked with 1.6-mm ball bearings and CT scanned. Bone surfaces were coated with 99mTc-doped paint. The locations of the ball bearings were marked with paint doped with 111In. Separate SPECT scans were acquired using the energy windows of 99mTc and 111In.

Results: Serial SPECT images aligned with an average root-mean-square (RMS) error of 3.8 mm (i.e., <1 pixel). CT-to-SPECT volume matching aligned with an RMS error of 7.8 mm. Total counts in CT-defined VOIs applied to SPECT data showed a strong linear correlation (r2 = 0.86) with true counts obtained from a dose calibrator.

Conclusion: The capability of this multimodality registration technique to anatomically localize and quantify radiotracer uptake is sufficiently accurate to warrant further assessment in an in vivo trial.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Bone Transplantation
  • Craniotomy
  • Humans
  • Image Processing, Computer-Assisted*
  • In Vitro Techniques
  • Skull / diagnostic imaging*
  • Skull / surgery
  • Tomography, Emission-Computed, Single-Photon*
  • Tomography, X-Ray Computed*