Linguistic Approach in Intelligent Image Content Analysis of Heart Vessel Structures

Clinical practice shows that, despite the frequent use of major computer-assisted diagnostic equipment, the obtained visualisations of the spatial reconstructions of specific body parts, later undergo only a rough qualitative assessment by the physician diagnostician, with no in-depth semantic analysis. Moreover, often a new method of medical visualisation does present the essence of a medical problem, but the physician watching such a new visualisation is unable to interpret it properly because, in simple words, there are so many things shown that he or she does not know what to look at. This shows that the actual progress in the practical application of 3D visualisations in medicine may depend on the progress in developing smart diagnostics support systems, making use of automatic analysis and understanding of medical 3D images. The problem tackled in this work is the semantic interpretation and an attempt at automatic understanding of a 3D structure of spatially visualised coronary vessels.