Search

Along with the massive influence of computational technologies in medical research and practice, the wide generation of patient, clinical and lab tests data make the assistance of intelligent information systems very important for correct therapy, surveillance and advising of the patient. Medical data is highly heterogeneous and with different structure and formats. Principal tasks of the work are focused at intelligent integration of all data related to a particular patient. Aside with the methodological aspects of linking medical data, an important part of the discussed study focuses on the design and development of a special purpose decision support system. This system enriches a specific ontology for diabetes mellitus with a set of newly defined rules, supporting the development of broad and more precise personalized dietary recommendations as a part of digitized healthcare assistance.

Contemporary development in personalized medicine based both on extended clinical records and implementation of different high-throughput “omics” technologies has generated large amounts of data. To make use of these data, new approaches need to be developed for their search, storage, analysis, integration and processing. In this paper we suggest an approach for integration of data from diverse domains and various information sources enabling extraction of novel knowledge in cancer studies. Its application can contribute to the early detection and diagnosis of cancer as well as to its proper personalized treatment.

Semantic enhancement links sections of data files with well-described concepts from some knowledge domain. This allows for further automated reasoning about that data and can be especially useful for extracting value from Big Data. Most of the available enhancement tools focus on specific enhancement needs and data types. In this paper we present our efforts to expand the GLOBDEF framework, introduced in an earlier work, which aims to find a way for processing of large amounts of data and enhancing the data automatically. The framework is designed to leverage a variety of external enhancement tools and has no limitations on the format of the enhanced data. We demonstrate how the framework behaves on a mixed data set of texts and images and explain how an image can be semantically enhanced with a simple automated combination of an object recogniser and a text-based automated enhancer.