Acoustic impedance interpretation of cross-sectional human skin by using time and frequency domain deconvolution

In the human cheek skin assessment using an ultrasound microscope, deconvolution is the process to estimate the reflection coefficient that comes only from the skin without the interference of any other components. The conventional deconvolution method which is performed only in the frequency domain will lead to the instability of the reconstructed signal, especially on the low-frequency components of the signal, causing the generation of artifacts when the result is converted into the acoustic impedance distribution. In order to deal with this, this paper presents a deconvolution technique in both the time and frequency domain. The essential low-frequency components are determined by the time domain deconvolution, which is then combined with the high-frequency components from the conventional frequency domain deconvolution method, and the cross-sectional acoustic impedance profile along the depth direction has been successfully observed.