Two strategies achieve real-time computation and maintain the good results of the best denoising methods in noise removal and edge preservation. After the identification of the denoising method that best fits the requirements for ultrasound images: (i) the development of a computationally optimised version of a selected denoising algorithm, exploiting HPC tools, CPUs and GPUs, and low-level programming languages; (ii) the design and implementation of a deep learning framework that learns to replicate the output of the selected denoising algorithm.