Radiant energy – Invisible radiant energy responsive electric signalling – With or including a luminophor
Reexamination Certificate
1999-08-18
2001-09-25
Hannaher, Constantine (Department: 2878)
Radiant energy
Invisible radiant energy responsive electric signalling
With or including a luminophor
C250S363070, C250S369000
Reexamination Certificate
active
06294788
ABSTRACT:
BACKGROUND
The present invention relates to the field of positron imaging, and more particularly to the field of positron emission tomography. The invention is also applicable to other fields in where it is necessary to estimate the contribution of randoms in data indicative of positron coincidence events.
Positron emission tomography (PET) is a branch of nuclear medicine in which a positron-emitting radiopharmaceutical such as
18
F-fluorodeoxyglucose (FDG) is introduced into the body of a patient. Each emitted positron reacts with an electron in what is known as an annihilation event, thereby simultaneously generating a pair of 511 keV gamma rays. The gamma rays are emitted in directions approximately 180° apart, i.e. in opposite directions.
A pair of detectors registers the position and energy of the respective gamma rays, thereby providing information as to the position of the annihilation event and hence the positron source. Because the gamma rays travel in opposite directions, the positron annihilation is said to have occurred along a line of response (LOR) connecting the detected gamma rays. A number of such events are collected and used to reconstruct a clinically useful image.
One factor which degrades image quality in PET imaging is random events. The 511 keV gamma rays generated by the positron annihilations are detected within a narrow coincidence timing window. Pairs of such gamma rays detected within this timing window are ordinarily considered to be coincident and are used to generate an image. However, some of these events result from what are known as random events. A random event is one in which a pair of gamma rays from two unrelated annihilation events are detected in coincidence. Thus, the acquired coincidence data includes both true and random events. Because the LORs for the random events do not represent actual positron annihilations, the randoms introduce noise into the acquired data, thereby degrading image quality.
Various techniques have been used to minimize the deleterious effects of random events. Because the number of randoms increases with the square of activity, one technique is to image at relatively low activity levels. While relatively fewer randoms are detected, an undesirable side effect of this technique is that fewer true coincidence events are available to generate the image.
Another technique for estimating the contribution of randoms is to delay the signal from one of the detectors by an amount longer than the coincidence timing window prior to applying the coincidence check. Due to the delay, events which are detected by a pair of detectors within the coincidence timing window (i.e., in coincidence) represent randoms. The collected events are rebinned and used to correct the acquired coincidence data. A particular drawback to such a delayed correction technique is that the rate of randoms collection is the same as that of the true event collection. This technique also has a deleterious effect on image noise characteristics.
Yet another technique is to determine the random coincidence rate based on the singles rates of the system's detectors and the length of the coincidence timing window. According to one technique, the system detectors have been treated as a plurality of virtual subdetectors, and the singles rate for each of the subdetectors has been measured. The singles rates for the various combinations of subdetectors has in turn been used to generate a randoms sinogram. One disadvantage to such a technique is that it is necessary to collect data additional to the desired coincidence data. Yet another disadvantage is that improving the accuracy of the estimation requires that the detectors be divided into arbitrarily small subdetectors.
SUMMARY
Aspects of the present invention address these matters, and others.
According to one aspect of the present invention, a method of position imaging includes receiving data indicative of a plurality of detected coincident gamma ray pairs, said pairs including positron annihilation gamma ray pairs and random gamma ray pairs, re-pairing gamma rays from the detected coincident gamma ray pairs so as to generate non-coincident gamma ray pairs, and using the coincident gamma ray pairs and the non-coincident gamma ray pairs to generate a randoms corrected image.
According to a more limited aspect, the step of receiving includes receiving a list of detected coincident gamma ray pairs and the step of re-pairing includes re-pairing gamma rays from the list of detected coincident gamma ray pairs.
According to another more limited aspect, step of re-pairing includes pairing each of a plurality of gamma rays with a non-coincident gamma ray. According to a still more limited aspect, the invention includes pairing each of a plurality of gamma rays with a plurality of non-coincident gamma rays. The non-coincident gamma rays may be paired in all possible combinations.
According to another more limited aspect of the present invention, the step of receiving includes receiving data indicative of coincident gamma ray pairs detected over a time period T. The method also includes rebinning the non-coincident gamma ray pairs and resealing the rebinned non-coincident gamma ray pairs to generate T*R pairs, where R is a randoms rate.
According to another limited aspect, the method includes combining the coincident and non-coincident gamma ray pairs and generating an image indicative of the combined coincident and non-coincident gamma ray pairs. The method may also include rebinning the coincident gamma ray pairs, rebinning the non-coincident gamma ray pairs, subtracting the rebinned non-coincident gamma ray pairs from the rebinned coincident gamma ray pairs to generate corrected coincidence data, and generating an image indicative of the corrected coincidence data.
According to still another limited aspect, the method includes using first and second detectors to detect coincident gamma ray pairs. The step of re-pairing includes pairing gamma rays detected by the first detector with non-coincident gamma rays detected by the second detector. According to a yet more limited aspect, the method method may include using first, second, and third detectors to detect coincident gamma rays and wherein the step of re-pairing includes pairing gamma rays detected by the first detector with gamma rays detected by the third detector and pairing gamma rays detected by the second detector with gamma rays detected by the third detector.
According to another aspect of the present invention, a method of randoms-corrected imaging includes detecting coincident gamma ray pairs, said pairs including positron annihilation gamma ray pairs and random gamma ray pairs, generating non-coincident gamma ray pairs, and using the coincident gamma ray pairs and the non-coincident gamma ray pairs to generate a randoms corrected image.
According to a more limited aspect, the step of generating non-coincident gamma ray pairs includes detecting a plurality of single gamma rays and pairing each of a plurality of the single gamma rays with a non-coincident single gamma ray. According to a still more limited aspect, the method may include detecting coincident gamma ray pairs for a time period T, establishing a randoms rate R, and detecting at least 2*T*R single gamma rays. According to a still more limited aspect, the method may include pairing each of a plurality of the gamma rays with a plurality of non-coincident gamma rays. According to a still more limited aspect, he method may include detecting coincident gamma ray pairs for time period T, determining a randoms rate R, and rescaling the non-coincident gamma ray pairs to generate T*R pairs. According to a still more limited aspect, the method may include rebinning the coincident gamma ray pairs, combining the rebinned non-coincident event pairs with the rebinned coincident gamma ray pairs to generate randoms corrected data, and generating an image from the randoms corrected data.
According to yet another limited aspect of the present invention, the step of detecting includes using first, second a
Cooke Steven E.
Matthews Christopher G.
Vesel John F.
Fry John J.
Gagliardi Albert
Hannaher Constantine
Marconi Medical Systems, Inc.
LandOfFree
Randoms correction in positron imaging does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Randoms correction in positron imaging, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Randoms correction in positron imaging will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2482704