Computing the mechanical index.

A computational nonlinear beam propagation model was used to compute the water path and in situ fields of a phased array transducer operating at 2 MHz. The computational source was matched to the transducer's z = 10 cm focal plane field. Subsequent computed propagations considered this source operating at source amplitudes up to 1.49 MPa in a water medium and in a tissue medium with an attenuation of 0.3 dB cm(-1) MHz(-1). The mechanical index was calculated in three ways based on these computations: extrapolated from one low amplitude water path propagation, extrapolated from a series of water path propagations using the existing Output Display Standard protocol, and directly from a series of tissue path propagations. These computed results suggest that extrapolation from derated measurements of a low level water path field can provide mechanical index estimates which progressively overestimate the in situ values. At the highest source amplitude considered, the linearly extrapolated mechanical index was 29% higher than the mechanical index computed by the tissue path propagations. The Output Display Standard protocol offered improved accuracy but consistently underestimated the in situ values. The maximum error for the Output Display Standard protocol was 8%. A variation of the Output Display Standard protocol in which mechanical index estimates were obtained from the on-axis spatial peak in the derated temporal peak rarefactional curve was also considered. The maximum error for this method was 3%. The results considered here also demonstrated the feasibility of computational investigations of high intensity clinical propagations.