X-ray or gamma ray systems or devices – Specific application – Mammography
Reexamination Certificate
2002-10-25
2004-10-12
Bruce, David V. (Department: 2882)
X-ray or gamma ray systems or devices
Specific application
Mammography
C250S363020
Reexamination Certificate
active
06804325
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to positron emission mammography, and more particularly to methods and apparatus for positron emission mammography image reconstruction.
BACKGROUND OF THE INVENTION
Functional imaging of the breast with F-18 fluorodeoxyglucose (FDG) has the capability to differentiate metabolically active tumors and is being investigated for the detection, staging and treatment of breast cancer. Clinical breast imaging with FDG has been performed with conventional positron emission tomography (PET) scanners as well as with dedicated positron emission mammography (PEM) devices. There have been many proposed designs for PEM detectors, which have the potential for improved sensitivity and spatial resolution compared with conventional PET scanners. Initial clinical results have been reported only for the dual planar detector configuration, however.
For a PEM device with the breast positioned between two static planar detectors, image reconstruction is usually performed by backprojection along the lines of response (LORs) to the desired image planes [18]. A lesion is focused when an image is formed in the plane in which it is located, while out-of-plane activity appears blurred. In this contribution three major factors affecting PEM image formation by the backprojection method are investigated: image uniformity (flood) corrections, image sampling (pixel size) and count allocation methods. In addition, enhanced PEM image reconstruction by iterative matrix methods is described.
Image uniformity is dependent on spatially dependent geometric and detector sensitivity factors.
In conventional PEM the source region or tissue with positron-emitting activity, for example F-18 fluorodeoxyglucose (FDG), is placed between two detectors operated in coincidence mode. The detectors are capable of detecting the two 511 keV photons resulting from positron annihilation following radioactive decay and positron emission. The detector heads are parallel to each other. Images of the source distribution can be formed by backprojecting events (counts) along the line of response (LOR) connecting the centers of the detection pixels in the two detector heads. In planes parallel to the detector heads, counts are allocated to the pixel through which the LOR passes. This method of count allocation is also sometimes termed the nearest neighbor or closest pixel approximation. The contribution of an event may be weighted using a detection probability (See C. J. Thompson, K Murthy, Y. Picard, I. N. Weinberg, and R. Mako, “Positron Emission Mammography (PEM): A Promising Technique for Detecting Breast Cancer,”
IEEE Transactions on Nuclear Science
, vol. 42, pp. 1012-1017, 1995).
This method of image reconstruction by backprojection is also known as laminography. In some implementations images are reconstructed using only those events whose raypaths are within a certain maximum angle from normal incidence with the detectors and without a uniformity correction (See R. R. Raylman, S. Majewski, R. Wojcik, A. G. Weisenberger, B. Kross, V. Popov, and H. A. Bishop, “The Potential Role of Positron Emission Mammography for Detection of Breast Cancer. A Phantom Study,”
Medical Physics
, vol. 27, pp. 1943-1954, 2000). These images exhibit non-uniformities in part because the number of lines of response through a pixel in the reconstructed image varies as a function of the pixel's spatial position. Uniformity corrections are sometimes applied using measured data from activity in a uniform box phantom taken at a single position between the detectors.
The spatial dependence of positron camera sensitivity has been previously noted and displayed in
FIG. 12
of Muehllehner et al. (See G. Muehllehner, M. P. Buchin, and J. H. Dudek, “Performance Parameters of a Positron Imaging Camera,”
IEEE Transactions on Nuclear Science
, vol. NS-23, pp. 528-537, 1976). The geometric efficiency of a circular planar positron camera has been studied by Weathersby et al. (See P. K. Weathersby, S. S. Survanshi, and P. Meyer, “Spatial Sensitivity of a Planar Positron Camera,”
Nuclear Instruments and Methods in Physics Research
, vol. 220, pp. 571-574, 1984). PEM detectors also have been developed by Irving Weinberg [1984] (See I. N. Weinberg, S. Majewski, A. G. Weisenberger, A. Markowitz, L. Aloj, L. Majewski, D. Danforth, J. Mulshine, K Cowan, J. Zujewski, C. Chow, E. Jones, V. Chang, W. Berg, and J. Frank, “Preliminary Results for Positron Emission Mammography: Real-time Functional Breast Imaging in a Conventional Mammographic Gantry,”
European Journal of Nuclear Medicine
, vol. 23, pp. 804-806, 1996), and in our laboratory (See R. R. Raylman, S. Majewski, R. Wojcik, A. G. Weisenberger, B. Kross, V. Popov, and H. A. Bishop, “The Potential Role of Positron Emission Mammography for Detection of Breast Cancer. A Phantom Study,”
Medical Physics
, vol. 27, pp. 1943-1954, 2000).
Nearest neighbor count allocation in backprojection reconstruction may result in banding or pixelization artifacts in the reconstructed image if the pixel size in the reconstructed image is smaller than the detector pixel size. Such artifacts have been seen in images reconstructed using PEM detectors built at Jefferson Lab. Banding and pixelization artifacts are visible in PEM image reconstructions in
FIG. 5
of K. Murthy, M. Aznar, C. J. Thompson, A. Loutfi, R. Lisbona, and J. H. Gagnon, “Results of preliminary clinical trials of the positron emission mammography system PEM-I: a dedicated breast imaging system producing glucose metabolic images using FDG,”
Journal of Nuclear Medicine
, vol. 41, pp. 1851-1858, 2000.
Freifelder and Karp in R. Freifelder and J. S. Karp, “Dedicated PET Scanners for Breast Imaging,”
Physics in Medicine and Biology
, vol. 42, pp. 2463-2480,1997 implemented iterative image reconstruction for numerical simulations of dual detector PEM, but only after rebinning the coincidence data using the single slice rebinning algorithm. Iterative reconstructions were performed for each slice separately, thus image reconstruction of the entire three-dimensional source region was not performed at the same time, which is desirable in order to obtain the most likely (in the statistical sense) estimate of source activity in the 3-D region between the detectors.
While all of the foregoing analysis and image data manipulation techniques provided useful and highly valuable information, certain image artifacts that resulted when the image reconstruction pixel size was smaller than the detector pixel tended to distort the image. In the case of back projection reconstruction, because of the “slice reconstruction” approach used, the lesion contrast tended to be somewhat reduced.
OBJECT OF THE INVENTION
It is therefore an object of the present invention to provide computer programs and formulae that enhance the quality of the PEM acquired image for more accurate detection and localization of a potential lesion
SUMMARY OF THE INVENTION
The method of the present invention uses list mode event data acquired from a dual head positron emission mammography detector or from numerical simulations of event data from such a detector. A computer program reads in the event data, performs the image reconstructions and writes out a binary file with the reconstructed image. This computer program can perform both backprojection image reconstruction and iterative image reconstruction. A copy of the source code to this program is attached hereto as Appendix A.
The image reconstruction method of the resent invention comprises accepting coincidence data from either a data file or in real time from a pair of detector hgeads, culling event data that is outside of a desired energy range, saving the disried data for each detector position or for each pair of detector pixels on the two detector heads, and the reconstructing the image either by backprojection image reconstruction or by iterative image reconstruction. In the backprojection image reconstruction mode, rays are traced between centers of lines of response (LOR's), a pro
Bruce David V.
Southeastern Universities Research Assn.
LandOfFree
Method for position emission mammography image reconstruction does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for position emission mammography image reconstruction, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for position emission mammography image reconstruction will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3319458