Radiant energy – Invisible radiant energy responsive electric signalling – With or including a luminophor
Reexamination Certificate
2001-04-30
2003-08-05
Gagliardi, Albert (Department: 2878)
Radiant energy
Invisible radiant energy responsive electric signalling
With or including a luminophor
C250S363090, C250S363070
Reexamination Certificate
active
06603125
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to the art of nuclear medicine and diagnostic imaging. It finds particular application in localizing a scintillation event in a gamma camera having a number of photomultipliers arranged over a camera surface. It is to be appreciated that the present invention may be used in conjunction with positron emission tomography (“PET”), single photon emission computed tomography (“SPECT”), whole body nuclear scans, transmission imaging, other diagnostic modes and/or other like applications. Those skilled in the art will also appreciate applicability of the present invention to other applications where a plurality of pulses tend to overlap, or “pile-up” and obscure one another.
Diagnostic nuclear imaging is used to study a radio nuclide distribution in a subject. Typically, one or more radiopharmaceutical or radioisotopes are injected into a subject. The radiopharmaceutical are commonly injected into the subject's bloodstream for imaging the circulatory system or for imaging specific organs that absorb the injected radiopharmaceutical. A gamma or scintillation camera detector head is placed adjacent to a surface of the subject to monitor and record emitted radiation. Each detector typically includes an array of photomultiplier tubes facing a large scintillation crystal. Each received radiation event generates a corresponding flash of light (scintillation) that is seen by the closest photomultiplier tubes. Each photomultiplier tube that sees an event generates a corresponding analog pulse. Respective amplitudes of the pulses are generally proportional to the distance of each tube from the flash.
A fundamental function of a scintillation camera is event estimation, which is the determination of energy and position of the location of an interacting gamma or other radiation ray based on the detected electronic signals. A conventional method for event positioning is known as the Anger method, which sums and weights signals seen by tubes after the occurrence of an event. The Anger method for event positioning is based on a simple first moment calculation. More specifically, the energy is typically measured as the sum of all the photomultiplier tube signals, and the position is typically measured as the “center of mass” of the photomultiplier tube signals.
Several methods have been used for implementing the center of mass calculation. With fully analog cameras, all such calculations (e.g., summing, weighting, dividing) are done using analog circuits. With hybrid analog/digital cameras, the summing and weighting are done using analog circuits, but the summed values are digitized and the final calculation of position is done digitally. With “fully digital” cameras, the tube signals will be digitized individually. In any event, because the fall-off curve of the photomultipliers is not linear as assumed by the Anger method, the image created has non-linearity errors.
One important consideration is the location of the event estimation. The scintillation light pulse is mostly contained within a small subset of the tubes on a detector. For example, over 90% of a total signal is typically detected in seven (7) out of a total number of tubes, typically on the order of 50 or 60. However, imaging based only on the seven (7) closest tubes, known as clustering, has poor resolution and causes uniformity artifacts. Furthermore, because the photomultiplier tubes have non-linear outputs, the scintillation events are artificially shifted toward the center of the nearest photomultiplier tube.
For a given detector geometry, the fall-off curve varies with a depth that a gamma photon interacts in the crystal. Different energy photons have varying interaction depth probabilities that are more pronounced in thicker crystals, which are typically used in combination with PET/SPECT cameras.
Therefore, separate linearity or flood correction tables are created and used for each energy in order to correct for the uniformity artifact. Fall-off curves are acquired using a labor intensive method of moving a point source a small amount (e.g., 2 mm) roughly 30-40 times for each tube. The individual tube's output is acquired at each location, the mean value of the tube's output is found, and a curve of tube output versus distance from the location of the point source is generated.
A disadvantage of generating a fall-off curve using a point source is the large amount of time required to move the source position. This method is also prone to errors in positioning the source accurately on the detector. It is also usually only done in one or two directions. Therefore, the assumption is made that the fall-off curve is exactly symmetric. Regenerating the fall-off curve for a different energy requires that the process be repeated again. Likewise, generating the fall-off curve for a different tube requires the process be repeated again. Therefore, the assumption is usually made that the fall-off curve is invariant across different detectors or photomultiplier tubes.
Generating the linearity correction tables typically involves using a lead mask that contains many small holes to restrict the incident location of radiation on the crystal surface. The holes represent the true location of the incident photons that interact in the detector crystal. This information is used to generate a table that consists of x and y deltas that when added to the x and y estimate, respectively, are used to generate a corrected position estimate that more accurately reflects the true position. A disadvantage is that new tables must be generated for each energy that is to be imaged, thereby increasing the calibration time. Another disadvantage is that the calibration mask has a limited number of holes, since each must be resolved individually, thereby limiting the accuracy of the correction. It is also increasingly more expensive and difficult to calibrate for higher energy photons since the thickness of the lead mask must increase in order to have sufficient absorption in non-hole areas.
Another prior art method uses separate flood uniformity correction tables for each energy. A disadvantage is that new tables must be generated for each energy that is to be imaged, which increases calibration time. Flood correction has the disadvantage of creating noise in the image, since the method is based on either adding or removing counts unevenly throughout the pixel matrix. This method is also sensitive to drift in either the photomultiplier tubes or electronics.
Another prior art method reduces the output from the closest tube. For example, an opaque dot is sometimes painted over the center of each photomultiplier tube. The sensitivity can also be reduced electronically. Unfortunately, the closest photomultiplier tube typically has the best noise statistics. Reducing its sensitivity to the event causes a resolution loss.
Similarly, excluding the outlying tubes reduces the noise in the determined values of energy and position. The most common way of excluding signals from outlying tubes includes imposing a threshold, such that tube signals below a set value are either ignored in the calculation or are adjusted by a threshold value. This method works reasonably well in excluding excess noise. However, the method fails if stray signals exist above the threshold value. Stray signals may exist at high-counting rates, when events occur nearly simultaneously in the crystal. When two events occur substantially simultaneously, their “center-of-mass” is midway between the two—where no event actually occurred. Nearly simultaneously occurring events may result in pulse-pile-up in the energy spectrum and mispositioning of events. This behavior is especially detrimental in coincidence imaging, where high-count rates are necessary.
Thus, it is desirable to improve localization in event estimation. With a fully digital detector, both the intensity and the location of each tube signal are known. It is, therefore, possible to calculate the energy and position based primarily on the tube signals close to an indiv
Cooke Steven E.
DiFilippo Frank P.
Vesel John F.
Fay Sharpe Fagan Minnich & McKee LLP
Gagliardi Albert
Koninklijke Philips Electronics , N.V.
LandOfFree
Event localization and fall-off correction by... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Event localization and fall-off correction by..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Event localization and fall-off correction by... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3075850