The influence of ignoring the acoustic attenuation parameter during full waveform inversion of transcranial ultrasound data – Preliminary Study

Full wave inversion (FWI) is a data inversion technique used extensively in seismology field to estimate physical properties of earth’s subsurface by minimizing the misfit between the observed data and the synthetic data. Only recently its use has been explored in the domain medical imaging, especially for transcranial ultrasound imaging. Although FWI gives better resolution compared to other computed tomography methods, it is computationally heavy. Moreover, there are no widely used standard software/packages to solve FWI problems in medical imaging. Here we explore the use of an open source toolbox based on MATLAB to perform our simulations. Further, we also explore the effect of ignoring attenuation in the backpropagation step of adjoint-state modelling in the FWI technique, and observe that the convergence is faster and the structural similarity index (SSIM) metric is marginally better when compared to adjoint-state modelling with attenuation.