3193
Physics in Medicine & Biology
Estimation of dynamic time activity curves
from dynamic cardiac SPECT imaging
J Hossain
1
, Y Du
1
, J Links
2
, A Rahmim
3
, N Karakatsanis
3
,
A Akhbardeh
4
, J Lyons
5
and E C Frey
1
1
Division of Medical Imaging Physics, Russell H. Morgan Department of Radiology
and Radiological Science, Johns Hopkins Medical Institutions, Baltimore,
MD 21287, USA
2
Division of Molecular and Translational Toxicology, Bloomberg School of Public
Health, Johns Hopkins University, Baltimore, MD 21205, USA
3
Division of Nuclear Medicine, Russell H. Morgan Department of Radiology
and Radiological Science, Johns Hopkins Medical Institutions, Baltimore,
MD 21287, USA
4
MRI Research, Russell H. Morgan Department of Radiology and Radiological
Science, Johns Hopkins Medical Institutions, MD 21287, USA
5
Department of Neurology, Brigham and women’s hospital department of neurology
and Harvard Medical School, Boston, MA 02115, USA
E-mail: jakirh2000@gmail.com
Received 25 August 2014, revised 26 January 2015
Accepted for publication 25 February 2015
Published 26 March 2015
Abstract
Whole-heart coronary low reserve (CFR) may be useful as an early predictor
of cardiovascular disease or heart failure. Here we propose a simple method to
extract the time-activity curve, an essential component needed for estimating
the CFR, for a small number of compartments in the body, such as normal
myocardium, blood pool, and ischemic myocardial regions, from SPECT data
acquired with conventional cameras using slow rotation.
We evaluated the method using a realistic simulation of
99m
Tc-teboroxime
imaging. Uptake of
99m
Tc-teboroxime based on data from the literature were
modeled. Data were simulated using the anatomically-realistic 3D NCAT
phantom and an analytic projection code that realistically models attenuation,
scatter, and the collimator-detector response. The proposed method was then
applied to estimate time activity curves (TACs) for a set of 3D volumes of
interest (VOIs) directly from the projections. We evaluated the accuracy
and precision of estimated TACs and studied the effects of the presence of
perfusion defects that were and were not modeled in the estimation procedure.
The method produced good estimates of the myocardial and blood-pool
TACS organ VOIs, with average weighted absolute biases of less than 5% for
the myocardium and 10% for the blood pool when the true organ boundaries
Institute of Physics and Engineering in Medicine
0031-9155/15/083193+16$33.00 © 2015 Institute of Physics and Engineering in Medicine Printed in the UK
Phys. Med. Biol. 60 (2015) 3193–3208 doi:10.1088/0031-9155/60/8/3193