jump to navigation

“Time-of-Flight” takes off in PET/CT October 5, 2007

Posted by tomography in CT, Nuclear Medicine, PET, Tomography, Uncategorized.

TOF or time-of-flight reconstruction is actually not a recent invention in tomography as its application in diagnostics was first proposed in the 1980s, but only recent technological improvements allowed its use in scanners. In a diagnostic scan improvement means better image quality, and less tracer dose for the patient.
In a PET scan when a positron collides with an electron, two gamma rays are generated in the annihilation process. The detector ring is looking for two, almost “simultaneous” rays, which is then noted as an event and is stored as data. Naturally, only those two rays are simultaneous, that traveled the same distance. Therefore using the time difference (fraction of a second) between the gamma rays, the scanner can pinpoint the original location of the positron-electron annihilation.tofschematic


To best understand it, you need to look at conventional PET imaging, which involves the injection of a decaying radioactive agent into a patient. As each nucleus decays, a positron is released that immediately collides with an electron, creating an annihilation that releases a pair of photons, or gamma rays. These two photons travel away from the collision point at 180° from each other. After detecting the photons, the PET scanner’s computer uses that information to calculate where the radioactive agent is concentrated and produce an image localizing the affected area. TOF makes it possible for point of origination of annihilation to be more accurately predicted, which leads to more accurate imaging. Improved event localization reduces noise in image data, resulting in higher image quality, shorter imaging times, and lower dose to the patient

A time-of-flight mass spectrometer uses the differences in transit time through a drift region to separate ions of different masses. It operates in a pulsed mode so ions must be produced or extracted in pulses. An electric field accelerates all ions into a field-free drift region with a kinetic energy of qV, where q is the ion charge and V is the applied voltage. Since the ion kinetic energy is 0.5mv2, lighter ions have a higher velocity than heavier ions and reach the detector at the end of the drift region sooner.

TOF is especially useful in examining overweight patients who have been traditionally difficult to image.



1. Scott Welpe, CNMT - October 6, 2007

Time of flight can not offer “reduced noise in image data, higher image quality, shorter imaging times and lower dose to the patient” in PET or PET•CT imaging. My rebuttal being based around the use of the word ‘and’. Time of flight will reduce image noise successfully but as soon as imaging times or dose are reduced the loss of image statistics/counts puts image quality back where you started. The issue with the current commercial offering time of flight is that the timing resolution that makes ToF possible increases to around 900 picoseconds from the stated 650 picoseconds. This reduces the expected image quality gain. The gain from ToF is seen in large patients but offers little to no value in head and neck imaging nor pediatric or small patients. Additionally there is no improvement in resolution in ToF, this would require about a 20 fold improvement in the current detector electronics to achieve this.

There are other commercial offerings that improve signal to noise and resolution like the use of point spread function (PSF). This approach matches the signal to noise improvements of ToF but is independent of dose and improves resolution. Traditionally resolution degrades as you move from the center of the field of view, PSF uniquely offers uniform resolution throughout the entire field.

2. Paul Michael Stanley - November 11, 2007


Your facts are very inaccurate. It sounds to me like you are quoting the timing resolution from the Discovery RX and it’s failed attemppt to implement Time-of-Flight. That is the system you would have experience with from Duke correct? You are not a Time of Flight user with first hand experience with the hardware are you?

Your data stating that commerical implementation of TOF yields more like 950 psec timing res. shows your unfamiliarity with the only commercial product supporting this capability. The timing resolution on the systems I am familiar with are nowhere near your stated figure…as a matter of fact the HUP system is reporting at 575 psec! It’s also not accurate to cite any NEMA published specs for the TOF based system as NEMA doesn’t have any TOF based spec measurements at this time.

Interesting though to hear you try and minimize the quality gain “even if it was 900 psec”. 1000psec would improve positional accuracy to a space of less than 12 cm. This is compared to EVERY OTHER PET SCANNERS capability of being able to position no better than 60 cm or the entire FOV. 12 cm vs 60 cm? C’mon? How can one minimize even that significant gain?

Please read Dr. Cherry’s “Of Mice and Men…..” article from JNM, Dec 2007. TOF – One of the 5 biggest advancements in PET Scan technology.

It’s very disappointing to know that this industry we work in is continuously being held back from applauding the advances in capability because of vendor competition. This same competition should strive to give our field more awarenes to the medical world thru innovation, but instead it forces people into downplaying what they themselves don’t have as being “not relevant”.

PSF is used in the SMS branded HD and while very beneficial for recon, how would you argue their claims to derive 2mm spatial resolution as a result of it? Very misleading to say the least, correct?
Your information about PSF is like reading a Siemens brochure…… PSF does not match signal to noise. PSF does not improve uniformity of FOV. Siemens increase in S2N is due to increased material usage and extended axial FOV via TrueV and their new claim to uniform FOV is due to TrueX not the PSF (aka HDPET).

Anyway, to not dilute my intent – all of the advancements in PET, whether it be HD, Vue Point, or TruFLight and Time of Flight are wonderful. None of them should be downplayed, but Time of Flight specifically is in a league of it’s own and represents what PET has been working toward demonstrating towards next generation capability.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: