Photoacoustic image quality enhancement by estimating mean sound speed based on optimum focusing

A constant sound speed (i.e., 1540m/s) is generally used in various ultrasound (US) and photoacoustic (PA) image reconstruction algorithms based on the assumption of a homogeneous sound speed in human tissue. However, the variation of the sound speed in human tissue can be as great as 10%, which can lead to low contrast, distortion, and blurring in reconstruction images. We proposed an automatic method of selecting the mean sound speed based on optimum focusing of received PA signals to enhance the quality of reconstructed PA images. Optimum focusing is quantified by calculating the minimum sum of the deviation of beamformed PA signals from their mean value for various sound speeds. The proposed method was demonstrated by homogeneous and heterogeneous sound speed simulation models, and also evaluated by experiments with agar and porcine tissue-mimicking phantoms. The central vertical and lateral profiles of reconstructed absorbers verified the improvement of contrast, signal-to-noise ratio (SNR), and spatial resolution by using the estimated mean sound speed. (C) 2015 The Japan Society of Applied Physics