PT  - JOURNAL ARTICLE
AU  - Nikolai Slavine
AU  - Roderick McColl
AU  - Padmakar Kulkarni
TI  - Iterative deconvolution method for pre-clinical PET/CT Image enhancement: application in imaging Alzheimer's plaque deposition in AD transgenic mice
DP  - 2016 May 01
TA  - Journal of Nuclear Medicine
PG  - 1985--1985
VI  - 57
IP  - supplement 2
4099  - http://jnm.snmjournals.org/content/57/supplement_2/1985.short
4100  - http://jnm.snmjournals.org/content/57/supplement_2/1985.full
SO  - J Nucl Med2016 May 01; 57
AB  - 1985Objectives To evaluate an efficient iterative method to improve the quality of sub-optimal pre-clinical PET/CT images.Even the state-of-the -art PET imaging device has limitations in detecting small disease foci in mice brains and in assessing plaque burden quantitatively. A big disadvantage is that 3D reconstruction time to obtain high quality images is too long and 3D reconstruction algorithm convergence is too slow. A novel iterative recovery method (RSEMD) to reduce noise and enhance resolution as compared with other conventional expectation-maximization (EM) algorithms has been demonstrated. Methods F-18 quinoline (an amyloid binding tracer) was used in imaging 12 mo. old APP/PS1 AD transgenic mice. In one study, dynamic PET images were obtained up to 30 min. immediately following the tracer administration and in another study, the animals were sacrificed 5 min post injection and imaged for 15 min.The RSEMD method was applied to noisy PET/CT images previously reconstructed with conventional EM method to determine improvements in SNR and CNR. The method was tested in progressive manner from validation in samples of small animal imaging studies of non-invasive detection of beta-amyloid plaque in transgenic mouse models of Alzheimer's disease from commercial Siemens Inveon® micro PET/CT scanner. Results Starting from raw data, 10-20 iterations were sufficient to obtain an appropriate quality image with low noise, high resolution and reduced reconstruction time (from hours to few minutes). In all of the small-animal PET/CT studies the post-processed images proved to have higher resolution and lower noise as compared with images reconstructed by conventional EM methods and without introducing additional artifacts or declining in spatial resolution. In general, the values of SNR reached a plateau at around 10-20 iterations with an improvement factor of about 2 for sub-optimal PET images. Conclusions A rapidly converging, iterative deconvolution algorithm with a novel resolution subsets-based approach RSEMD that operates on DICOM images has been developed. The RSEMD method has promising potential to enhance suboptimal PET/CT images by post-processed image recovery to diagnostically acceptable level. This approach may be suitable for quantitative measurement of changes in Alzheimer's pathology in experimental animal studies in noninvasive assessment of efficacy of novel therapeutic interventions. $$graphic_6D99876B-8EAA-41EF-A885-40345EFB2390$$