Estimation of dynamic time activity curves from dynamic cardiac SPECT imaging

Area of Science:

• Nuclear Medicine
• Cardiovascular Imaging
• Medical Physics

Background:

• Whole-heart coronary flow reserve (CFR) is a valuable early predictor of cardiovascular disease and heart failure.
• Estimating CFR requires accurate time-activity curves (TACs) from dynamic SPECT imaging.
• Conventional SPECT imaging methods may face challenges in precise TAC extraction, especially with complex physiological conditions.

Purpose of the Study:

• To develop and evaluate a simple method for extracting TACs from dynamic SPECT data.
• To assess the accuracy and robustness of the proposed TAC estimation method in the presence of myocardial perfusion defects.
• To enable more reliable CFR estimation for early cardiovascular disease detection.

Main Methods:

• A novel method was proposed to extract TACs for myocardial and blood pool regions from slow-rotation SPECT data.
• Realistic simulations using the 3D NCAT phantom and (99m)Tc-teboroxime uptake models were employed.
• The method was applied to estimate TACs from simulated projections, evaluating accuracy and precision with and without modeled perfusion defects.

Main Results:

• The method yielded good estimates for myocardial and blood-pool TACs with low bias (<5% and <10% respectively) when organ boundaries were known and activity was uniform.
• TAC estimation remained accurate (average weighted absolute bias <10%) even with unknown perfusion defects totaling ≤5% reduction in uptake.
• Accurate TACs were obtained for normal and defect regions when larger defects (25% extent, 100% severity) were included in the estimation procedure (average weighted absolute bias ≈5%).

Conclusions:

• The proposed method provides accurate and robust TAC estimation from dynamic SPECT data, even with moderate perfusion defects.
• This technique shows promise for improving CFR quantification and aiding in the early diagnosis of cardiovascular conditions.
• The method's robustness to model mismatch enhances its clinical applicability in real-world SPECT imaging scenarios.