Ultrasound is one of the most commonly used noninvasive medical imaging techniques. Ultrasound contrast agents (UCA), consisting of encapsulated gas-filled microbubbles, have shown to increase the diagnostic precision in selected low echogenic patients. UCA also holds promise for bedside evaluation of myocardial perfusion quantification, but is not yet reproducible and specific enough for clinical use. In addition risks have been addressed when used, as first recommended, together with high mechanical index (MI) for reperfusion assessment by contrast destruction. We clinically observed increased myocardial velocities after UCA-administration when applied simultaneously with color tissue Doppler imaging (CTDI) arising the question if this increase was due to physiological factors or physical changes in the backscattered signals when UCA were present.

The aims of the thesis was to explain this velocity shift and simultaneously to contribute to a future safe and contrast specific application by further characterizing the non-linear acoustic properties of UCA when located in an acoustic field. Of specific interest was to evaluate in which way nonlinear wave propagation affects the response from UCA and if a change in pulse shape, length or polarity can be utilized to increase the nonlinear signal contribution.

Twelve patients with ischemic heart disease were examined with CTDI before and after UCA-administration in order to verify the change in peak systolic velocity. An experimental in vitro model including flow and tissue phantoms for UCA was established for CTDI. Raw data from single-element transducers and clinical ultrasound systems were collected for three different UCA and analyzed to determine if the observed velocity shift could be reproduced in vitro and to find a possible cause. Our results show in vivo and in vitro that UCA will affect the autocorrelation phase shift estimator used for CTDI in terms of contribution from rupturing UCA microbubbles, which explains the velocity shift. CTDI during contrast infusion should therefore be avoided unless it can be performed at low MI where the majority of the UCA are intact.

The computational model for spatial superposition of attenuated waves was modified to include an operator for pulse distortion from nonlinear wave propagation. The Matlab™ toolbox Bubblesim based on a modified Rayleigh-Plesset-equation and with insonation parameters such as frequency, pressure amplitude, pulse length and polarity was used to study the response from single microbubbles either for simulated pulses or for pulses generated by clinical ultrasound systems and single element transducers. The combination of the two models also provided a computational platform to asses pulse distortion from nonlinear wave propagation, the response of the UCA bubble and the linear backscatter of the low amplitude bubble echo. When evaluating the harmonic response in simulations and in vitro, the interaction of the excitation pulses with the contrast bubbles was identified as the main cause of nonlinear scattering, and a 2-3 dB increase of the second harmonic amplitude depends on nonlinear distortions of the incident pulse. By applying small changes of short (<3.5 cycles) and fragmented transmitted wideband pulses of 2-2.5 MHz, it is shown that inverted pulse polarity considerably modulates power without affecting a low and safe MI (<0.4), and the results lodged promise to further to enhance a contrast response.