Contrast-to-tissue ratio improvement by transmitted optimized binary signal in ultrasound pulse inversion imaging
Résumé
Ultrasound contrast imaging has provided more accurate medical diagnoses. One of the most used techniques is the pulse inversion imaging improving the contrast-to-tissue ratio (CTR) by extracting nonlinearities of contrast agents. The usual transmitted signal is at a fixed frequency. However, an optimal choice requires information about the transducer and the medium. This information is experimentally inaccessible. Moreover, the digital electronic setup can limit the solution. Our goal was to seek the binary command which maximized the CTR. A genetic algorithm sought the vector of input binary samples. By adding a closed loop, the system automatically proposed the optimal binary command without any a priori information about the system or the medium explored and without hypothesis on binary samples. In simulation, the gain compared with a transmitted signal at the optimal frequency can reach about 3 dB and 0.5 dB in comparison with a transmitted signal at the two-thirds of the central frequency of the transducer.
