Trade-off between preamplifier noise figure and decoupling in MRI detectors

There is a limit to the maximum achievable preamplifier decoupling. In many cases, this level is not enough. To overcome this limit, the preamplifier noise figure can be compromised for further decoupling increase. This is useful in flexible MRI arrays where ensuring coil insensitivity to changes in other array elements is a challenge.

This work establishes the relation between the preamplifier noise figure and preamplifier decoupling using closed-form equations. These equations allow the evaluation of preamplifier decoupling properties and benchmark different preamplifiers against each other. The method to design the corresponding decoupling networks is described. The derived generalized design equations, which are not limited to 50 Ω pre-matched preamplifiers, greatly improve design flexibility and enable use of new amplifiers in MRI detectors.

Using the method, the decoupling properties of three preamplifiers are studied. For demonstration, the coil decoupling is further increased by 10.8 dB using one of the preamplifiers. The noise figure is sacrificed by 0.5 dB, which is predicted by equations and verified experimentally. Although examples are shown for 3 T systems at 32.13 MHz and 127.7 MHz, the approach and equations apply to any field strength and nucleus.

Preamplifier decoupling can be improved beyond what is possible by traditional approaches. The derived design equations cover a wide range of cases, including inductive coils and self-resonant low-impedance and high-impedance coils.