出版社：Faculty of Humanities, Kaunas University of Technology
其他摘要：In this article we describe automatic image processing method for determination twist gray matter and some dark features of human brain in computer tomography images. Suggested algorithm of the method used six different functions for image processing. The first was noise-reducing function, which made possible to apply ”Roberts Cross” and matching functions. Threshold function separated the features of components of images into two values – black and white. In this way the amount of information for further processing was reduced. Dilation and erosion functions performed human brain twist matter selection and elimination of some dark images features that could be the ischemic stroke features. In Vilnius University Hospital ”Santariškės” 60 automatically processed computer tomography images of human brain were evaluated by experts in radiology. More than 90 % of processed images were evaluated in marks 10 and 9 using ten point’s evaluation system. Ill. 13, bibl. 8 (in Lithuanian; summaries in Lithuanian, English and Russian).
