Classification methods in image analysis with a special focus on medical analytics

This paper describes the design and application of classification methods for image analysis and processing. Accordingly, the main trends and challenges of the machine learning are presented in multiple contexts where the image analysis plays a very important role, including security and biometrics, aerospace and satellite monitoring, document analysis, natural language understanding, and information retrieval. This is accomplished by introducing a categorisation of the most challenging classification methods according to the thematic context and classification typology. Hence, supervised and unsupervised classification methods are presented and discussed. It is followed by a special focus on the medical context, where the classification methods for image analysis are of prior importance in supporting the medical diagnosis process. Accordingly, the second part of the paper surveys the recent and current research in medical analytics where the image classification is a key aspect, and tracks the horizon of the research for future challenges in the field.