Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Order statistic-neural network hybrid filters for gamma camera-Bremsstrahlung image restoration

Journal Article · · IEEE Transactions on Medical Imaging (Institute of Electrical and Electronics Engineers); (United States)
DOI:https://doi.org/10.1109/42.222667· OSTI ID:6318307
; ;  [1]
  1. Univ. of South Florida, Tampa, FL (United States). Dept. of Radiology
A new class of filters, an order statistic and neural network hybrid filter (OSNNH), is proposed for the restoration of gamma camera images, based on the measured modulation transfer function. This filter shares the advantages of both neural network for deconvolution and advanced nonlinear filtering for noise removal and edge enhancement. Planar images of [beta]-emitting radionuclides are used to evaluate the new filter because they exhibit higher degradation than images of single photon emitters due to increased photon scattering and collimator septal penetration. The filter performance is quantitatively evaluated and compared to that of the Wiener filter by investigating the relationship between the externally measured counts from sources of phosphorus-32 ([sup 32]P) at various depths in water. An effective linear attenuation coefficient for [sup 32]P is determined equal to 0.13 cm[sup [minus]1] and 0.14 cm[sup [minus]1] for the OSNNH and the Wiener filters, respectively. Evaluation of phantom and patient filtered images demonstrates that the OSNNH filter avoids ring effects caused by the ill-conditioned blur matrix and noise overriding caused by matrix inversion, typical of other restoration filters such as the wiener. Although further work is required on the optimization of the OSNNH filter, the present results suggest that this new restoration method may be suitable for quantitative system calibration using [beta]-emitters, important for antibody therapy management, and for broader applications to single-photon studies.
OSTI ID:
6318307
Journal Information:
IEEE Transactions on Medical Imaging (Institute of Electrical and Electronics Engineers); (United States), Journal Name: IEEE Transactions on Medical Imaging (Institute of Electrical and Electronics Engineers); (United States) Vol. 12:1; ISSN 0278-0062; ISSN ITMID4
Country of Publication:
United States
Language:
English