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

This Article Abstract Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Riddell, C. Articles by Bacharach, S. L. Search for Related Content PubMed PubMed Citation Articles by Riddell, C. Articles by Bacharach, S. L.

Noise Reduction in Oncology FDG PET Images by Iterative Reconstruction: A Quantitative Assessment

National Institutes of Health, Bethesda, Maryland

View larger version (65K): [in a new window]   FIGURE 1. Mean of reconstructed image replicates for OSEM and FBP with and without AC for 4 different slices (A, B, C, and D). Negatives in FBP images are not displayed. For every slice, FBP-AC and OSEM-AC images are displayed to 1 common maximum, as are FBP-NAC and OSEM-NAC images. Hot tumor images have been clipped for display. Last row contains corresponding reconstructed attenuation maps with lung, heart, and liver regions of interest (ROIs). Whereas columns A, B, and C are displayed in standard supine position, breast images in column D are displayed in prone position.

View larger version (111K): [in a new window]   FIGURE 2. SD images displayed in same order as in Figure 1. Intensity of each pixel indicates magnitude of noise at that pixel. For every slice, FBP-AC and OSEM-AC images are displayed to 1 common maximum, as are FBP-NAC and OSEM-NAC images.

View larger version (107K): [in a new window]   FIGURE 3. S/N images displayed in same order as in Figure 1. Intensity of each pixel reflects intensity of S/N at that pixel. Images are clipped so that all can be displayed on 1 common scale (0–2.2).

View larger version (66K): [in a new window]   FIGURE 4. S/N percentage improvement (%I) images. First row compares FBP-AC S/N images (Fig. 3, row 1) with OSEM-AC S/N images (Fig. 3, row 2). Second row compares FBP-NAC S/N images (Fig. 3, row 3) with OSEM-NAC S/N images (Fig. 3, row 4). Light blue background indicates that FBP and OSEM had comparable S/N. Red-to-yellow areas are where OSEM was better than FBP. Dark blue–to–black areas are where FBP was better than OSEM.

View larger version (23K): [in a new window]   FIGURE 5. (A) Emission distribution (segmented slice of anthropomorphic torso phantom study) used for simulating emission replicate sinograms at different count levels. (B) Plot of average S/N with respect to count level in reconstructed OSEM images in heart cavity, lungs, hot background (Bck), and heart wall of simulated distribution. Lines through data points are fits of data to straight line.