HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Maass Moreno, R. Articles by Chen, C. PubMed Articles by Maass Moreno, R. Articles by Chen, C.

Post-reconstruction motion correction in PET using point-sources

¹ Nuclear Medicine, National Institutes of Health, Bethesda, Maryland; ² Radiology, UCSF, San Francisco, California; ³ Radiation Oncology, National Cancer Institute, Bethesda, Maryland

241

Objectives: A method is presented to remove motion blur on PET (or SPECT) images.

Methods: The method exploits the unambiguous location and intensity that a set of point-sources produce on an image and that, when attached to a moving patient, allow the identification of detectable discrete motions. The relative intensity of the markers measures the contribution of each patient position to the blurred image. A patient-motion model is estimated from the markers’ intensities and, using a 'procrustes' algorithm, from their successive locations. A de-blurring procedure is then applied: i) iteratively via the difference between the observed image and the guessed image blurred with the model, or ii) analytically calculating the inverse of the model. To verify the method, four 185 kBq, 22Na point-sources (~.8 mm3) were attached to a phantom (200 nBq/ml). Motion-blurred images were created by moving the phantom during acquisition or by adding two or more images of a single phantom image previously rotated and translated via software.

Results: Iterative solutions converged accurately to the un-blurred image but not in every case. Analytic solutions produced accurate and more robust results. Inaccuracies in the model are identifiable with at least one of the solutions and interactive fine-tuning was possible.

Conclusions: This method does not require list-mode acquisition or light-tracking devices. Iterative (when converged) and analytical solutions provided accurate blur-free PET images. The next step will be to test, on real patient data, whether the procedure improves diagnostic confidence or increases target delineation accuracy for radiation therapy.

Research Support: In part by the NIH,CC Intramural Research Program

This Article Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Maass Moreno, R. Articles by Chen, C. PubMed Articles by Maass Moreno, R. Articles by Chen, C.