Deep learning approaches to biomedical image segmentation

Abstract

The review considers automatic image segmentation using deep learning methods in the field of medical imaging. Current developments in the field of machine learning, especially those related to deep learning, are useful for identifying and quantifying models in medical images. The main point of these achievements is the significant ability of deep learning approaches to obtain hierarchical representations of characteristics directly from images, which, in turn, eliminates the need for manually created functions. Deep learning is rapidly transforming into state-of-the-art medical imaging technology and has led to improved productivity in various clinical applications. This review discusses the basics of deep learning methods, as well as an overview of successful implementations that include image segmentation for various medical applications. Finally, some research questions are highlighted, and the need for further improvements is indicated in the future.