Spatial Resolution calculation using wavelength (medical imaging)

[nobody0]

Hi
Im comparing PET scans and MRI scans and I've noticed that MRI
s spatial resolution is 0.3-1mm while PET scans have a spatial resolution much larger, around 5-7mm.

I was just wondering how these figures are calculated. I've looked up some formulae and all I could find was wikipedia, with a formula for telescopes. In particular I need a formula which will make this work in PET and MRI calculations.

What I can't get my head around is that PET use gamma rays which have a shorter wavelength, which it would logically seem would produce more detail (a smaller/better spatial resolution) than the radio waves MRI uses (larger wavelength).

Is there something I am missing here?

Thanks

 

[JeffKoch]

Hi
Im comparing PET scans and MRI scans and I've noticed that MRI
s spatial resolution is 0.3-1mm while PET scans have a spatial resolution much larger, around 5-7mm.

I was just wondering how these figures are calculated. I've looked up some formulae and all I could find was wikipedia, with a formula for telescopes. In particular I need a formula which will make this work in PET and MRI calculations.

What I can't get my head around is that PET use gamma rays which have a shorter wavelength, which it would logically seem would produce more detail (a smaller/better spatial resolution) than the radio waves MRI uses (larger wavelength).

Is there something I am missing here?

Thanks

I'm not familiar with medical imaging equipment specifically, but generally speaking without knowing a lot of details about the systems, there is no formula because the resolution limit (actually it's not a number, it's a curve like a point-spread function or a modulation transfer function) of any imaging system depends on a number of factors. Wavelength is often not the most important factor, you need to consider what element does the imaging and how it works, how well does it image (for example even in a telescope the angular resolution is rarely determined solely by diffraction, it's also determined by alignment, mirror/lens quality, atmospheric turbulence, etc.), what feature contrast you're trying to see, detector spatial resolution and detected photon statistics, background statistics, and so on.

So there's no simple answer to your question, except to go to the manufacturer of whatever instruments you're using and see if they can provide you with something useful. I use x-ray imaging systems all the time for other applications, but the spatial resolution I get depends entirely on all these details, and might vary from 5-500 microns, vastly larger than the wavelengths used - I could do better but only in some specialized applications, and I generally don't have the photon statistics to make use of it anyways. If I was imaging a static object at a synchrotron, the game would change entirely, and so would all the numbers.

 

[maimonides]

The optical resolution limit (in the telescope/microscope formula) only applies to devices operating on wave optics principles. MRI and PET don´t.