Iterative Image Processing for Early Diagnostic of Beta-Amyloid Plaque Deposition in Pre-Clinical Alzheimer's Disease Studies.

Abstract

Purpose: To test and evaluate an efficient iterative image processing strategy to improve the quality of sub-optimal pre-clinical PET images. A novel iterative resolution subsets-based method to reduce noise and enhance resolution (RSEMD) has been demonstrated on examples of PET imaging studies of Alzheimer's disease (AD) plaques deposition in mice brains.

Materials and methods: The RSEMD method was applied to imaging studies of non-invasive detection of beta-amyloid plaque in transgenic mouse models of AD. Data acquisition utilized a Siemens Inveon® micro PET/CT device. Quantitative uptake of the tracer in control and AD mice brains was determined by counting the extent of plaque deposition by histological staining. The pre-clinical imaging software inviCRO® was used for fitting the recovery PET images to the mouse brain atlas and obtaining the time activity curves (TAC) from different brain areas.

Results: In all of the AD studies the post-processed images proved to have higher resolution and lower noise as compared with images reconstructed by conventional OSEM method. In general, the values of SNR reached a plateau at around 10 iterations with an improvement factor of about 2 over sub-optimal PET brain images.

Conclusions: A rapidly converging, iterative deconvolution image processing algorithm with a resolution subsets-based approach RSEMD has been used for quantitative studies of changes in Alzheimer's pathology over time. The RSEMD method can be applied to sub-optimal clinical PET brain images to improve image quality to diagnostically acceptable levels and will be crucial in order to facilitate diagnosis of AD progression at the earliest stages.