In classical MRI phased-array design, optimal coil overlapping is used to minimize coupling between nearest-neighbor coils, and low input impedance preamplifiers are used to isolate the relatively weak coupling between non-nearest neighbors. However, to make the complex sensitivities of phased-array coils sufficiently distinct in parallel spatially-encoded MRI, it is desirable to have no overlapping between coils. Also, if phased arrays are used as transmit coils in MRI, one can no longer rely on the low input impedance of the preamplifiers for decoupling. Here a coupling and decoupling theory is introduced to provide a better understanding of the relations between coupled and uncoupled signals in the MRI phased array, and to offer a new method for decoupling phased-array coils without overlapping the nearest coil pairs. The new decoupling method is based on the assumption that any n-element phased array can be decoupled by a 2n-port interface system between phased array and preamplifiers. The detailed analysis and the experimental results show that a four-port interface can be used to decouple a two-element phased array. Furthermore, the four-port interfaces can serve as building blocks to construct a 2n-port decoupling interface. This new method allows one to place the coil elements anywhere that could optimize parallel spatial encoding without concern for coupling between the coils. The method can also be used for phased-array transmit coils.
Copyright 2002 Wiley-Liss, Inc.