Comparison of ordered subsets expectation maximization and Chang's attenuation correction method in quantitative...

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Radiological Physics

Background:

• Photon attenuation significantly degrades cardiac single photon emission tomography (SPET) image quality, causing artifacts.
• Accurate attenuation correction is crucial for reliable quantitative analysis in cardiac SPET.

Purpose of the Study:

• To compare the quantitative accuracy and performance of the ordered subsets expectation maximization (OSEM) reconstruction algorithm versus Chang's non-uniform attenuation correction method in cardiac SPET.
• To evaluate the impact of these methods on defect quantification and image uniformity.

Main Methods:

• Cardiac SPET scans were performed on an anthropomorphic phantom simulating myocardial studies.
• Attenuation maps were derived from computed tomographic images.
• SPET projection data underwent attenuation correction using both OSEM reconstruction and Chang's method, followed by automated 3D quantitation of defect parameters.

Main Results:

• Quantitative differences between OSEM and Chang's method for defect-to-normal ARM ratio, maximum defect contrast (MDC), and defect volume were within 4%, 19%, and 13%, respectively.
• Calculation times for both methods were comparable.
• OSEM demonstrated a more accurate radiotracer distribution and improved uniformity compared to Chang's method, with segmental analysis showing less than 8% difference in uniformity for normal studies.

Conclusions:

• OSEM and Chang's attenuation correction methods are quantitatively equivalent for cardiac SPET, with similar computational demands.
• OSEM reconstruction yields a more accurate representation of radiotracer distribution, making it the preferred algorithm for cardiac SPET.
• The findings support the use of OSEM for improved image quality and quantitative accuracy in cardiac SPET.