Simulation study of strain imaging using nonlinear ultrasound radio frequency signals

Ultrasound based strain imaging is a promising non-invasive method to diagnosis tumors or cancers since tissue stiffness indicates the physiological as well as pathological states of living bodies. But linear ultrasound wave theory based strain images suffer from low signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), poor spatial resolution and image artifacts. Applications of nonlinear ultrasound theories on strain estimation might be useful to improve the image quality. This paper presents a simulation study on strain images constructed using nonlinear ultrasound wave with both fundamental and second harmonic radio frequency data. The finite element method (FEM) is used to model tissue deformation and then the corresponding pre- and post-compression radio-frequency (RF) signals are generated using MATLAB based ultrasound simulation programs. The strain was calculated using analytic minimization (AM) of regularized cost functions and Kalman filter reducing the decorrelation noise. The obtained strain images were compared with images obtained from the linear ultrasound RF signals. Simulation results show that nonlinear methods can generate strain images with higher SNR, CNR and resolution compared to the linear method.