alexa Investigation of attenuation correction in SPECT using textural features, Monte Carlo simulations, and computational anthropomorphic models.
Chemical Engineering

Chemical Engineering

Journal of Analytical & Bioanalytical Techniques

Author(s): Spirou SV, Papadimitroulas P, Liakou P, Georgoulias P, Loudos G

Abstract Share this page

Abstract PURPOSE: To present and evaluate a new methodology to investigate the effect of attenuation correction (AC) in single-photon emission computed tomography (SPECT) using textural features analysis, Monte Carlo techniques, and a computational anthropomorphic model. MATERIALS AND METHODS: The GATE Monte Carlo toolkit was used to simulate SPECT experiments using the XCAT computational anthropomorphic model, filled with a realistic biodistribution of (99m)Tc-N-DBODC. The simulated gamma camera was the Siemens ECAM Dual-Head, equipped with a parallel hole lead collimator, with an image resolution of 3.54 × 3.54 mm(2). Thirty-six equispaced camera positions, spanning a full 360° arc, were simulated. Projections were calculated after applying a ± 20\% energy window or after eliminating all scattered photons. The activity of the radioisotope was reconstructed using the MLEM algorithm. Photon attenuation was accounted for by calculating the radiological pathlength in a perpendicular line from the center of each voxel to the gamma camera. Twenty-two textural features were calculated on each slice, with and without AC, using 16 and 64 gray levels. A mask was used to identify only those pixels that belonged to each organ. RESULTS: Twelve of the 22 features showed almost no dependence on AC, irrespective of the organ involved. In both the heart and the liver, the mean and SD were the features most affected by AC. In the liver, six features were affected by AC only on some slices. Depending on the slice, skewness decreased by 22-34\% with AC, kurtosis by 35-50\%, long-run emphasis mean by 71-91\%, and long-run emphasis range by 62-95\%. In contrast, gray-level non-uniformity mean increased by 78-218\% compared with the value without AC and run percentage mean by 51-159\%. These results were not affected by the number of gray levels (16 vs. 64) or the data used for reconstruction: with the energy window or without scattered photons. CONCLUSION: The mean and SD were the main features affected by AC. In the heart, no other feature was affected. In the liver, other features were affected, but the effect was slice dependent. The number of gray levels did not affect the results. This article was published in Nucl Med Commun and referenced in Journal of Analytical & Bioanalytical Techniques

Relevant Expert PPTs

Relevant Speaker PPTs

Recommended Conferences

Relevant Topics

Peer Reviewed Journals
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals
International Conferences 2017-18
Meet Inspiring Speakers and Experts at our 3000+ Global Annual Meetings

Contact Us

© 2008-2017 OMICS International - Open Access Publisher. Best viewed in Mozilla Firefox | Google Chrome | Above IE 7.0 version