Talk:Positron emission tomography
|This article is of interest to the following WikiProjects:|
|Ideal sources for Wikipedia's medical content are defined in the guideline Wikipedia:Identifying reliable sources (medicine) and are typically review articles. Here are links to possibly useful sources of information about Positron emission tomography.
- 1 Are
- 2 PET results
- 3 New image
- 4 Data units for PET Images
- 5 Antimatter?
- 6 Kinetic Analysis of PET-CT scan
- 7 PET Reconstruction
- 8 Annihilation in body
- 9 Annihilation after some time?
- 10 Conservation of Momentum
- 11 Lead Paragraph Quibble
- 12 Brand names in illustration
- 13 Slight Problem
- 14 Duplication
- 15 Query -- scope of a PET scan?
- 16 Safety
- 17 Resource "seeing pain"
- 18 fluorodeoxyglucose
- 19 History section missing Michel Ter Pogossian
- 20 Add an image for the direction of detection
- 21 Ref 12 is a dead link
- 22 Proposed merge
- 23 Refs 22 & 39 are identical
- 24 Cost per scan
Are the scan times quoted for PET/CT machines correct? The machine I have experience of - it's a couple of years old - requires significantly longer - the CT part takes approximately 10 to 14 minutes for a H&N CT, the slow time mainly being a result of the fact that the CT is very much a low-grade CT scanner. I am not sure what other machines do, and obviously times are very much dependent on the protocol being used, area being imaged, etc., but (in my experience) the CT part takes significantly longer than 30 seconds. zebdee
Answer: With todays multi-detector helical CT scans, the scan time is much faster. We can do an "eyes to thighs" CT scan in less than a minute. NucsDoc
- And one should add in the context of combined PET/CT that this is always the case, since these are always the newer scanners and have the latest CT features, which are always VERY fast. The 1-minute full CT scan is typical there. It's the PET part that takes 10 or 15 minutes a section and will be much harder to speed up with future tech (though of course it will happen). SBHarris 09:33, 4 November 2006 (UTC)
What are the percentages of false positives for PET scans?
Answer: PET is not specific, it merely tells you which areas of the body have faster metabolism than others. This means it picks up inflammation, infection and injuries as well as tumors. For this reason, PET is only suitable for diagnosing tumors in conjunction with other tests. Its other main role is staging of tumors i.e. assessing the extent of a tumor that has already been found by other means.
For example, a CT scan of the chest finds a single nodule in one lung. The question is then, "is this cancer?". PET scanning in this situation can tell you whether this nodule is likely, or unlikely to be tumor.
There was a recent review of PET imaging in lung cancer. In the case of the solitary pulmonary nodule scenario, the specificity of a "positive" PET scan was 82%-94% depending on the precise PET technique used.
For staging of lung cancer - the most important role for PET is determining whether a lung tumor has spread to the center of the chest (the mediastinum). However, even in this case there is a significant false positive rate (PPV : 74-93%; false positive rate: 44.6%). There is also a significant false-negative rate in this situation (up to 11.7%), so PET is usually used in conjuction with other tests, including exploratory surgery (mediastinoscopy). ChumpusRex 15:04, 5 March 2006 (UTC)
Hello, I come from french Wikipedia, I upload a new image for oncology part. Enjoy !
Data units for PET Images
PET Images are measured in different units like Bacqurel per mili litre, SUV (Standard Uptake Value), Percentage and Counts/sec What does each of these units mean?
The problem with this unit is that the value depends on how much radioactive tracer was given to the patient, the size of the patient, how much time has passed since the injection (due to radioactive decay), etc.
The SUV is a method for partially compensating for this. It corrects for the total amount of radioactivity given, and the approximate volume of the patient (volume is very difficult to measure, so it is estimated from weight).
In words, the SUV is the uptake in the tissue divided by an estimate of the concentration in the body that was available to access the tissue. Concentration is approximated as the injected activity divided by body weight. The idea is to normalize the counts in the tissue to the counts that were available to be taken up by the tissue, or to "standardize" the uptake. Markssss (talk) 04:52, 28 July 2010 (UTC)
There are a variety of other methods used for calculating SUV, some correct for blood glucose concentration, some use lean body weight instead of total body weight, some estimate body surface area.
Percentage is a cruder method of correction. The voxel with the highest activity is found, and it's activity is called 100%. The activity in other regions is then presented as a fraction of that.
There still some debate as to whether the SUV is a useful measurement, from the point of diagnosis. Some abnormal areas (e.g. tumors), will have a higher uptake (and hence SUV) than normal tissues, and studies have suggested that the higher the activity of an abnormality, the more likely it is to be signficant. Some studies have looked at 'threshold' values of SUV, below which an abnormality is unlikely to be significant (or above which an abnormality is likely to be significant). E.g. if PET can be used to assess a nodule in the lung. If the SUV of the nodule is higher than 2.5, some studies have suggested that it's likely to be cancer.
However, given the different interpretations of SUV, different scan techniques, different times between injection and scanning at different hospitals, etc. some people argue that the SUV is relatively useless - as it's whole point is to allow comparisons between scans done at different times and at different sites.
ChumpusRex 21:43, 10 June 2006 (UTC)
I was just curious why there is a link to antimatter in the "See also" section. I couldn't see how the two were related, so I think there needs to be some mention of antimatter in the article if it relates, or the link should be deleted. cøøkiə Ξ (talk) 04:13, 22 September 2006 (UTC)
- Positrons are antimatter. This is clear if the wiki link to positron is followed. However, I see your point that it would be sensible to mention this in the article. ChumpusRex 12:14, 22 September 2006 (UTC)
Kinetic Analysis of PET-CT scan
right now i am working on a project.the topic is 'Development of a parameter which will distinguish between the TB n cancer' and this has to be done by kinetic analysis of PET-CT scan.
I haven't seen anything in this page related to the reconstruction process, which is a very important component of PET, because of the low data quality of the process. Do you think that's worth including here? User:Sjayanthi November 3rd, 2006
- Absolutely, if you know something about it. If it's arcane, you can put it in its own section toward the end of the article, as something like "Image reconstruction techniques important to PET" and reference it higher in the section which first mentions how the scanner opperates in more general terms. Actually, there are many ways of doing this. The main article on PET scanning should probably not be too much longer, but long sections on ANY subtopic in PET scanning can be summarized in their own short sections, with a main article referenced there. Then, you can put in a lot more detail in that main article. This is how Wikipedia expands. The main rules here are: don't delete, improve! Next is BE BOLD in adding stuff. The worst that will happen if you know what you're talking about, (provided you don't hit a malicious delete-man), is that your section will be edited down, and the extra material offloaded to a subarticle. Don't worry too much about writing quality-- we need good information worst. Writing can be improved by secondary writers who have a lesser understanding of material. So go for it. SBHarris 09:14, 4 November 2006 (UTC)
Annihilation in body
I'm not much of an expert in these fields, being a year 9 science student, however in the article under the sub-heading 'How the scanner operates', it reads "the positron travels a few millimetres before encountering and annihilating with an electron." Could somebody clarify for everybody (especially me) which material (the body or any other material) the electron is a part of? Because I had always thought that an annihilation of a sub-atom in the body would result in the atom concerned becoming a positively charged ion, and causing disruption in the atomic structure of the organs being built from these atoms. Thanks. 18.104.22.168 09:28, 19 November 2006 (UTC)
- Electrons wiped out by such positrons come from any old atom out there, and they do indeed become cations (positive ions) as a result. However, we're not talking about large numbers compared normal amounts of cations in the body (look at all those Na+,K+ and Mg++ floating around in all your fluids!). So it's no problem. If you calculate the number of moles of ions created from a normal PET scan in an hour, it's tiny-- something on the order of 10^12, which is a picomole compared with millimolar concentrations of your normal cations. Or 100 nanocoulombs with assocated currents of 50 picoamps, compared with microamps produced in your heart and brain.
Far, far worse damage comes from the gamma, which is ionizing also, but the damage is not caused by the ion per se (it's called that only because this is how these things were DETECTED originally), but rather from the high energy interactions which provide enough energy to break DNA. A simple ion contacting DNA won't do that. DNA is ionized (lots of negatives) and associates with positively ionized histones to be stored. Lots of ions in water are not bad, and in biology are normal. By the way, the atom the positron come from becomes a negative ion also, but this is soon fixed, as it picks up an H+ from solution and becomes a sugar hydroxyl. Again, no sweat, and generally no damage (due to the very small amounts of stuff made). SBHarris 08:47, 16 November 2006 (UTC)
Great, thanks a million. A+, here I come! 22.214.171.124 09:28, 19 November 2006 (UTC)
There is another interesting potential decay pathway of the positron - instead of annihilating with a free electron, the positron may strip off and then bind with (by coulomb attraction) a weakly bound electron, with the consequent formation of positronium (Ps), which is also sometimes referred to as "a light hydrogen atom". Ps can exist in either of 2 quasi-stable states - a triplet spin 1 state (3S1) which eventually decays by annihilation with a spin opposed electron from a neighboring atom in a process known as "pick-off annihilation", or in a singlet spin 0 state (1S0) which eventually decays by self combination. Note that with the decay of the spin 0 state, the process occurs in approximately 100 ps with two gamma photons being created and emitted at 180 degrees, while the spin 1 state decay occurs much more slowly in approximately 100 ns, and creates 3 gamma photons. (see http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=14761021&query_hl=3&itool=pubmed_docsum and http://www.ph.unimelb.edu.au/photo/people/leighton/thefinalcheck.pdf) Duedilly 06:12, 11 December 2006 (UTC)
Annihilation after some time?
i just want to know about the line that says "the positron travels a few millimetres before encountering and annihilating with an electron." but i thought that as soon as a positron encounters an electron, it annihilates...and electrons are present anywhere and everywhere.so how is that positrons are able to travel upto distances as large as milimeters inside the body...please somebody explain this...urgent ,specialy for me...i have a related project even...
- The distance any GIVEN particle travels in ANY material is related to its energy (all other things being equal), due to the fact the short wavelength particles, which of course are high energy particles, just "look" smaller, so have a smaller cross section. High energy electrons go a couple of mm in skin for radiotherapy, and Mev electrons up to several cm. A positron's no different. SBHarris 16:08, 28 March 2007 (UTC)
Conservation of Momentum
Hi I have a quick question about why the gamma rays emmitted by the positron-electron annihilation does not produce two rays exactly 180 degrees apart? Doesn't this break the law of conservation of momentum? Or does the initial velocity of the positron towards an electron before they annihilate cause the final gamma rays to not be 180 degrees apart? Thanks Ongy 09:41, 28 August 2007 (UTC)
You guessed it, the momentum of the electron and positron at the instance of annihilation causes the angle to not be an exact 180 degrees. The angle is usually close enough to 180 and the diameter of scanner bore is short enough that it doesn't considerably skew localization of the annihilation event. 14:51, 31 August 2007 (UTC)
Thanks for that!!! Ongy 00:49, 8 September 2007 (UTC)
- Yes, in short, if the initial momentum of the positron toward the electron it was to annihilate was zero, the angle would be 180 degrees exactly. But remember that these positrons have all just been shot out of a nucleus in beta decay, so they're moving at quite a clip compared to the electrons around them. SBHarris 07:24, 5 May 2008 (UTC)
Lead Paragraph Quibble
Hello, in the lead paragraph it says PET produces images of "metabolic activity" or something along those lines a couple of times. While this is the most popular use of PET, it is not necessarily the only use correct? PET can be used to image any process providing you can make the right biomolecule into a positron emitter. It is true that FDG is by far the most widely used clinically, and it images metabolism, but theoretically PET can image anything. Is this too specific for the lead paragraph? I understand the desire to state the most popular use, but as it is worded I think it is a bit deceiving. CallipygianSchoolGirl (talk) 02:27, 11 March 2008 (UTC)
You are absolutely correct, which is why I added something to that effect in the "Radioisotopes" paragraph some time ago. It should be said, however, that not just any molecule can be radiolabeled and turned into a successful PET probe. For example, there are some pretty severe constraints in developing ligands that reversibly bind to neuroreceptors in the brain. The isotope of choice is usually C-11, and so the compound must be able to be synthesized, injected in the subject and reach and pass its peak uptake all in a relatively short time, due to the short half life. Additionally, as the ligand gets metabolized in the liver, the metabolites that contain the C-11 cannot cross the blood brain barrier, or they will create a confound in the analysis of the kinetics. The result is many failed research tracers, but there are many successful ones as well. Markssss (talk) 05:57, 5 May 2008 (UTC)
Another quibble: The lead states that the PET scan and CT image are done at the "same time." Correct me if I'm wrong but they are still two seperate scans that do not overlap temporally. They are done in one sitting, but it is not like the CT X-Ray tube is firing at the same time the PET scanner is picking up Gamma is it? CallipygianSchoolGirl (talk) 02:31, 11 March 2008 (UTC)
Brand names in illustration
I notice that the illustration showing how the process works shows the Sun Microsystems brand name prominently, which bugs me. Would it be possible to redo this picture and remove this text and logo to make it generic and to remove the plug for the computer manufacturer? I'm sure Sun isn't the only one making this equipment. +ILike2BeAnonymous (talk) 20:51, 12 March 2008 (UTC)
- On the subject of bugs, I can't find Sun Microsystems mentioned anywhere in any of the illustrations; I did, however, remove a reference to Siemens in a caption. - Neparis (talk) 03:19, 15 March 2008 (UTC)
Query -- scope of a PET scan?
I was under the impression that a PET scan analyzes the patient's entire body, to find areas where the uptake of the tracer indicates possible cancer. A friend of mine has the impression that the scan is like an x-ray, in that it covers only a particular small segment of the body. Perhaps both types of scans are used, depending on the circumstances? I don't see the answer to this question in the article. My guess was that, to the knowledgeable editors working on the article, the answer was so obvious that it didn't occur to anyone to spell it out! JamesMLane t c 04:30, 22 October 2008 (UTC)
James, a PET scan is used to 'give an indication' whether an abnormal lump in the body is or is not an active 'cancer'. It isnt diagnostic. Any cell that takes up a lot of labelled glucose because it is rapidly metabolising sugar will show up as a bright spot on the PET scan. Your brain and liver, for example, use huge amounts of glucose all the time, consequently, both show up as positive on a PET scan. Unfortunately, when a part of the body has an active infection, the area becomes flooded with 'macrophage' cells normally - these are fighting the infection, and perfectly healthy, but will show up as PET positive. The theory goes that 'cancer' cells grow more rapidly than 'normal' cells, so take up more glucose than those surrounding them. In theory, therefore, if you have a CT scan of (say) the lung, and there is an abnormal lump in it, you cannot say whether it is a benign harmless lump or a cancer (unless you are treated on the NHS in the UK, and the MD you see chooses to save money and just guess). You need to actually grab a piece of the 'lump' and examine it under a microscope (ie have a biopsy and tissue diagnosis performed). If you have a CT scan with a lump in it, and then perform a PET scan, you end up with two superimposeable images: if the lump corresponds in space to an area on the PET scan that appears to be PET positive, you have an indication that the lump is undergoing rapid cell division (ie is cancerous) OR that the body has swamped the area with macrophage cells to fight a localised infection: so all in all, PET scans are not 100% diagnostic of cancer - you still need a biopsy. All in all, as they are currently used by HHS practitioners in the UK, PET scans are almost worthless, and tend to be used as an alternative to what NICE and SIGN guidelines describe as the 'gold standards' for diagnosis (ie physical biopsy and immunohistochemical laboratory work), in order to hide a) the general shortage of surgeons to perform biopsies because their waiting lists are too long, and b) the appallingly low standards of NHS Pathology laboratories, many of which lack full CPA Accreditation (Governments ' minimum Quality Control Standards) due to atrocious funding deficits. DrLofhouse —Preceding unsigned comment added by 126.96.36.199 (talk) 00:54, 29 January 2009 (UTC)
- The axial field of view (FOV) is the measure of what "length" of the body is being imaged at once in PET/CT, when you're doing just the PET part. That's about 8 inches right now, according to the literature http://jnumedmtg.snmjournals.org/cgi/content/meeting_abstract/48/MeetingAbstracts_2/46P-a. That means every 8 inches of your body must travel through the "ring" for a certain time, collecting gamma scientillations. I've been told (can't swear) that some of the newer scanners have 2 blocks of these 8 inch rings in a row, to give you more like 16 inches of height being counted at a time, so a "whole body" (they never count the body below the thighs-- nothing to see there usually) can be counted in just 3 sections. Usually a whole body PET does you from "eyes to thighs" that way, and takes 20 min to 40 min. The CT part is done in less than a minute and is over with, before you know it. It's the PET collection of radiation from the F-18 that takes all the time. And yes, usually if they'd bothered to get you into the machine and injected with isotope, they're not going to collect PET info for just a transverse slice 9 inches (or 18 inches) long! They usually go for the whole body, which is pelvis to neck, or might also include head. SBHarris 01:28, 29 January 2009 (UTC)
Under safety the article says that a PET scan is non-invasive. Yet the proecdure requires the injection of a radioactive substance on a biologically active molecule. Doesn't that make the procedure invasive?VR talk 16:54, 12 February 2010 (UTC)
It is noninvasive in the sense that it shows internal processes in living animals without surgically exposing the tissues. This is true both in the research and clinical settings. In research, this allows chemical processes to be quantified without having to dissect tissues in order to measure tracers within them. In clinical applications, this means, for example, that tumors can be identified and localized without exploratory surgery. You are correct in the sense that there are some very small risks involved, but obviously not the same as surgical procedures. Markssss (talk) 04:44, 28 July 2010 (UTC)
http://www.euronews.net/2011/11/23/x-rays-see-pain/ 23/11/11 13:07 CET
- Don't suggest merges to long articles. This article is already long, and if we merge all that stuff in, we'd just be under the gun to summarize out the fludeoxyglucose stuff again to a subarticle, per WP:SS. If you think there's some good stuff in the FDG article that needs summary HERE, feel free to do it, per WP:SOFIXIT. The balance between where info should go, and to what level of detail in main and subarticles is always tricky and subject to debate, and nobody will object if you tastefully copy stuff, or summarize stuff to here. Just so long as you don't MOVE anything wholesale, so it disappears THERE and REAPPEARS here. That's always bad when your target article is long, and the stuff you pilfer from is shorter. For example, the image you mention was once part of this article, and somebody felt the need to MOVE it to the FDG article, thus leaving nothing HERE. See the problem? I'll go and copy it, and leave a copy there. BTW, I agree with your taste, and like that image so much that it's been on my userpage for years! SBHarris 20:19, 11 July 2012 (UTC)
History section missing Michel Ter Pogossian
The current history section contradicts some facts, as apparently the first practical models were developed by Ter Pogossian and his team. From the NY Times obit: "Dr. Michel Mathew Ter-Pogossian, who led the research that turned the positron emission tomography (PET) scanner from an intriguing concept to a medical tool used in hospitals and laboratories everywhere... in the search for a workable model, great strides had been made by others over the decades, but it was Dr. Ter-Pogossian and his colleagues in St. Louis who produced a series of machines for commercial production and practical use in the 1970's... Early in his professional life, Dr. Ter-Pogossian built one of the first scanners to detect radioactivity concentrations in living material, as well as a tool for delivering radiation therapy for cancers of the cervix and uterus. In the early 1950's he recognized the potential of short-lived radioactive tracers, and he was one of the first to use them to find brain tumors and measure blood flow." http://www.nytimes.com/1996/06/21/us/michel-m-ter-pogossian-71-led-research-on-pet-scanner.html Thinkfood (talk) 07:21, 12 July 2012 (UTC)
Add an image for the direction of detection
Can you add an image like http://hyperphysics.phy-astr.gsu.edu/hbase/nucene/imgnuk/petscang.gif or http://www.whatisnuclearmedicine.com/upload/PET-SCAN-Brain.jpg ? It wasn't obvious to me that each detector can pick up activity within a cone, not a cylinder. In other words, each tube detector can see the entire head, not just a single point in the head. Otherwise it could only pick up events that occur in the very center of the circle, and I thought maybe the ring of detectors was translated around to scan other points in the head, or that there were multiple rings each focused on a different point, but now I see that's not how it works. — Preceding unsigned comment added by 188.8.131.52 (talk) 15:53, 8 October 2012 (UTC)
Moved to the end
- To clarify, this is the ref in the Operation subsection of Descriptions for the phrase which reads:
"during which time it loses kinetic energy, until it decelerates to a point where it can interact with an electron."
SBaker43 (talk) 23:32, 5 June 2013 (UTC)
The following articles are proposed for merge;
PET-CT has been proposed as a merge to this article for over a year, but this page hasn't been tagged. I have made this change and also propose that PET/MRI be merged here. This is because I believe the content would be better covered here, as the technology is very similar, as are the indications and outcomes. I think that coverage on this one page would improve the quality of the intended content of all three articles. LT90001 (talk) 09:30, 30 August 2013 (UTC)
- OPPOSE If we were to merge, should we merge PET/MR with PET, or with MR? A similar argument follows for PET/CT. As this is a rather niche field, in the case of PET/MR at least, I think it is better to keep separate articles. On an administrative note, it is much easier to consider the two proposed merges separately. GyroMagician (talk) 12:31, 4 January 2014 (UTC)
- OPPOSE I agree it should be kept in it's niche discussion. The technology is vastly different than PET and even quite different than PET/CT, specifically in the advantages of MRI (motion, soft tissue) and the challenges of AC (Attenuation Correction). While it appears to be a similar overall concept and technology, application of this technology has made it clear that it is vastly different. — Preceding unsigned comment added by Kkadams115 (talk • contribs) 17:06, 16 January 2014 (UTC)
- Support Think this would be a very good move Iztwoz (talk) 20:41, 26 October 2013 (UTC)
- Oppose - These are different, related topics best covered in individual articles. The descriptions of PET, MRI, and CT in the combined procedure articles should indeed be brief and link back to the main articles for more detail but this does not mean a merge is warranted. VQuakr (talk) 18:02, 16 January 2014 (UTC)
- Oppose - there is plenty to be said that is specific to PET/CT in its own right, to warrant a specific article about it. Inclusion of that in the PET article makes it over-long. I came to the PET/CT article to find out exactly what the /CT part meant - I already knew a little about PET. The article helped me. This is absolutely useful encyclopaedic information, deserving its own article. Of course, all 3 articles can be considerably improved - but a merger will not help that. Instead, the PET/CT and PET/MRI should concentrate on the specifics of those, to avoid excess overlap. Looking further, I see there are plenty of entire books, and thousands of papers, and of course a great many magazine pieces that are specifically about PET/CT. I have no doubt there is plenty of information in reliable sources to support the separate article. 184.108.40.206 (talk) 04:58, 2 March 2014 (UTC)
- Oppose Agree so precisely with reasoning immediately above, that I won't repeat it. Leave summaries of PET-CT and PET/MRI here, if you want something here, and have the others as main articles, per WP:SS. That is done here (in this article) already, please notice. This section can be expanded a bit if you want more summary. But don't shoe-horn in stuff you'll just have to spin off again (as it expands) later. SBHarris 03:08, 25 March 2014 (UTC)
Refs 22 & 39 are identical
- Fixed. If you see how I did that, you can do it yourself next time. You just name the reference something with a <ref name="NameHere"> tag, with full reference as usual and </ref> tag. Then, when you need to refer to the same reference again, you put in the front tag and no reference body, and add the end </ref> tag. You can put in the full reference name and tag anywhere in the article, as long as it's once and no more. The other tags are as described. SBHarris 03:31, 25 March 2014 (UTC)
Cost per scan
Cost per scan gives a figure of $1,100 for a PET scan which is woefully low and very misleading despite being properly sourced to a Cancer Care Ontario publication. I think this figure may be the cost to the government itself. In the U.S., I have seen many PET scan bills from multiple hospitals with several insurance co.'s as well as a few for individuals without insurance. The lowest by far was $2,000 for a bill with a 50% discount. However, this is an outlier and the cost typically is $4,000 to $6,000 to the insurance co. Unfortunately, my data falls under the heading of original research. Regardless, something needs to be done to not seriously mislead the reader as to the cost.TL36 (talk) 02:56, 2 April 2014 (UTC)