[ROBERT]

What is the limit of resolution of a 1.5 Tesla MRI bodyscan ?,
i.e. how big (area) would an object have to be to be detectable
on a bog standard 1.5 Tesla NMRI bodyscanner ?.

(By objective testing if possible, rather than manufacturer's "specifications").

[daveshorts]

I have no direct evidence of this, but I have have heard of fairly standard MRI scanners having a resolution of about 1mm³, although you would need several voxels (3D pixels) to be sure it was an object. However this is very dependent on how quickly you scan.

An MRI works becasuse hydrogen nuclei emit and recieve radio waves at different frequencies in different magnetic fields. so what you do is apply different magnetic fields in different places (these coils shaking is why the machines make so much noise) and throw radio waves in and listen for the echo coming back.
 what a scanner actually has is a limit on magnetic field resolution, the bigger the big field (the 1.5T one) is the better this is. So for a given amount of time you can scan a small space very well or a bigger one less well.

 There will also be an absolute limit, probably set by how sharply you can change the magnetic field per cm, and imperfections in the machine.

I hope that was interesting if not directly useful

[ROBERT]

Thank you Dave,
I believed that the resolution of a 1.5 Tesla MRI was about one square millimeter.
If this is correct then, an old fashioned X-ray would provide a much higher resolution image when imaging bone than a MRI, (although an X-ray would involve a dose of ionising radiation). 

[daveshorts]

Yes an X-ray is a higher resolution. The other advantage is that the contrast in MRI is a lot better between different kinds of soft flesh. An X-Ray is very good at looking at bones, but MRI works better for anything non-bones related.