In this talk, we will cover several of the numerical methods that have been explored to calculate ultrasound propagation through the skull bone. Numerical modeling has played a critical role to make therapy of brain disorders with focused ultrasound a reality. Because its acoustic mismatch with soft tissue and highly heterogeneous composition, the skull bone is a physical barrier that distorts ultrasound propagation, which makes challenging to ensure correct focusing of ultrasound at a clinical target in the brain. The combination ultrasound phase arrays with hundreds of independent elements, volumetric medical imaging such as computed tomography (CT), and numerical methods based on Finite-Difference Time-Difference (FDTD) to solve the equation of acoustic propagation established the first evidence that the distortion in ultrasound propagation caused by the skull barrier can be compensated. The numerical methods are used to calculate the phase signal that needs to be applied in each of transducer elements to counter the distortion introduced by the skull. However, these numerical methods to solve the full-wave equation, and despite essential advances in more efficient techniques such as spectral decomposition and use of graphic-processor units, remain computational intense and limit their applicability in a clinical setting. Currently, thalamotomy based on Magnetic Resonance Imaging-guided Focused Ultrasound uses pre-treatment CT data and simple ray theory calculations to program the therapy phase arrays. This simplistic approach, but computational efficient, is precise enough only for targets in the thalamic volume. To extend the treatable regions in the brain there is a critical need to continue to improve the methods that ensure the precision of full-wave solutions, but that yet remain highly efficient to be compatible with an operation in the clinic.