Medical Ultrasound Image Restoration in Presence of Defective Transducer Elements

Quality of medical ultrasound images is negatively affected by common defective elements in the ultrasonic transducer arrays. In this work, radio frequency (RF) signals from weak elements with low sensitivity are restored using an adaptive nonblind signal deconvolution method. For this purpose, the electromechanical impulse response of each element in the transducer must be measured. Initially, the negative effects of the RF signals recorded from each element are removed by Wiener deconvolution of RF signals through the measured impulse responses. Then to restore the RF signals, the deconvolution result is convolved with the reference electromechanical impulse response, corresponding to the intact and ideal element. To present the performance of this method, the electrical fatigue in piezoelectric materials, as a common real example for defective elements, is simulated by reducing the piezoelectric constant and permittivity parameters based on finite element method (FEM). The electromechanical impulse responses for the intact and fatigued transducers are obtained from the simulation. Finally, the quality of reconstructed point spread function (PSF) from the transducers is evaluated. The side lobe level (SLL) and full width at half maximum (FWHM) in axial and lateral directions are measured for the reference PSF from intact transducer and also original and compensated PSFs from the fatigued transducer. The results show that the metrics are improved to an acceptable level compared to the reference PSF after applying the proposed compensation method.