Ultrasound images provide important auxiliary information for clinical disease detection and diagnosis. The application of machine learning in ultrasound images has brought new changes to the analysis and diagnosis of ultrasound images. This paper first introduces the research status of ultrasound images from the aspects of denoising, segmentation, detection and classification of ultrasound images, and then summarizes the research and application progress of machine learning algorithms in the field of ultrasound images in recent years. Finally, this paper points out the difficulties and challenges faced by the research institute in the field of medical imaging based on machine learning, and looks forward to the future research direction, in order to provide reference and research ideas for further research and application of machine learning in the field of medical imaging.