Segmentation of gated Tl-SPECT images and computation of ejection fraction: A different approach

Citation
P. Brigger et al., Segmentation of gated Tl-SPECT images and computation of ejection fraction: A different approach, J NUCL CARD, 6(3), 1999, pp. 286-297
Citations number
30
Categorie Soggetti
Cardiovascular & Respiratory Systems
Journal title
JOURNAL OF NUCLEAR CARDIOLOGY
ISSN journal
10713581 → ACNP
Volume
6
Issue
3
Year of publication
1999
Pages
286 - 297
Database
ISI
SICI code
1071-3581(199905/06)6:3<286:SOGTIA>2.0.ZU;2-B
Abstract
Background. We describe a set of image processing algorithms and mathematic al models that can be advantageously used in schemes for the segmentation o f thallium-201-single photon emission computed tomography (SPECT) images an d for computation of left ventricular ejection fraction (EF). Methods. The system consists of two independent blocs for image segmentatio n and computation of function. The former is based on a multiresolution ell iptical coordinate transformation and dynamic contour tracking. Computation of EF is formulated on the basis of both the endocardial and epicardial co ntours, and we compare this formulation with that using only the endocardia l border for images with low signal-to-noise ratios,The accuracy of border detection was validated against manual border tracing on FDG-PET images, si mulated Tl-201-SPECT images where the true underlying borders were known, a nd actual Tl-201-SPECT images. Finally, we compared EFs computed for FDG-PE T, technetium-99m-SPECT and Tl-201-SPECT with those obtained from planar ga ted blood pool imaging. Results. The automatically obtained results always were within the manual u ncertainty range. Agreement between myocardial volumes from positron emissi on tomography and automatically obtained values from the simulated Tl-201-S PECT images was excellent (r = 0.95, n = 32), Agreement between EFs from pl anar gated blood pool imaging and the other image modalities was good (FDG- PET: y = 5.89 + 1.21x, r = 0.92, see = 6.24, n = 19, Tc-99m-SPECT: y = -3.8 6 + 1.06x, r = 0.88, see = 7.78, n = 9, Tl-201-SPECT: y = 17.8 + 0.81x, r = 0.77, see = 7.44, n = 26), For noisy input data the combined use of inform ation from epicardial and endocardial contours gives more accurate EF value s than the traditional formula on the basis of the endocardial contour only . Conclusions, Alternate approaches for segmentation and computation of funct ion have been presented and validated. They might also be advantageously in corporated into other existing techniques.