Tissue characterization: Influence of ultrasound setting on texture features in vivo

In diagnostic ultrasound, the echo graphic B-scan texture is an important area of investigation since it can be analyzed to characterize the histological state of internal tissues. An important factor requiring consideration when evaluating ultrasonic tissue texture is the ultrasound setting parameters. The aim of this study is to investigate the effect of varying the gain and the dynamic range setting on the texture features in-vivo using a 3D ultrasound probe. The left leg medial head of the gastrocnemius muscle of 5 healthy subjects were scanned. A region of interest was defined on each slice. Texture features including first order statistics, second order statistics, autoregressive model and wavelet transform parameters were extracted from the images. One-way ANOVA was used to test the significant difference between the varying settings. Varying gain setting yielded a large variation on the first order statistics (CV ranging from 41.18% for skewness to 86% for kurtosis) and less variation on second order statistics, autoregressive model and wavelet (CV ranging from 1.34% for autoregressive model to 35.03% for gradient). Most of texture features did not show a significant difference when considering the gain varying form 75 dB to 85 dB except gray level, variance and run length. Varying the dynamic range from 70dB to 90 dB yielded a small variation on all texture features (CV less 7%) indicating that changing the dynamic range has no significant affect on texture features.