MRI 1.5T vs 3T

Comparing 1.5T vs 3T MRI systems identifies in a number of differences:

Signal to noise ratio

Theoretically signal increases proportional to the square of the static field strength whereas noise increases linearly. This implies that in a perfect system SNR of a 3.0T system would be twice as good as at 1.5T. In reality, due to increase in susceptibility effects in most tissues the actual improvement is only in the 30-60% range (instead of 100%). With this increased SNR, the spatial resolution and/or acquisition time can be improved depending on which is more important for the particular case.

Specific absorption rate (SAR)

SAR is defined as the amount of radio frequency energy (joules) deposited tissues (kg). The limit set by the FDA is an amount which results in increase of 1 degree centigrade in any tissue 2. SAR is proportional to the static field (B0) to the power of 2, meaning a 3.0T system and 4 times as much as a 1.5T system. Additionally SAR is proportional to:

B0 squared
pulse duration and length
pulse number
slice number
flip angle
The last of the above, results in standard spin echo sequences which have 90 degree flips depositing to much energy. As a result there is a greater use of gradient echo sequences. Unfortunately these are imaging T2* not T2 and therefore are more susceptible to local field artefacts. Largely, these problems have been overcome with modern units.

Acoustic noise

Rapid gradient switching leads to increase in the intensity of the acoustic noise, which has required better insulation of both the unit itself and the room.