CASPIAN JOURNAL
MANAGEMENT AND HIGH TECHNOLOGIES
A survey of segmentation methods for medical mrt-images of brain tumors
 |
Read | Abdulraqeb Atef Rawhan Abdulsameea , Sushkova Lyudmila T., Lozovskaya Nina A. A survey of segmentation methods for medical mrt-images of brain tumors // Caspian journal : management and high technologies. — 2015. — №1. — pp. 122-138. |
Abdulraqeb Atef Rawhan Abdulsameea - post-graduate student, Vladimir State University, 87 Gorkov St., Vladimir, 600000, Russian Federation, atef_alsanawy@mail.ru
Sushkova Lyudmila T. - D.Sc. (Engineering), Professor, Vladimir State University, 87 Gorkov St., Vladimir, 600000, Russian Federation, ludm@vlsu.ru
Lozovskaya Nina A. - Ph.D (Medical), head of radiodiagnostic department; undergraduate, Aleksandro-Mariinsky Regional Clinical Hospital; Astrakhan State University
Segmentation of brain tumors on MRT-images is one of the difficult challenges in the field of digital medical image processing, since the location, shape and size of brain tumors are unpredictable factors. The presence of multiple segmentation algorithms described in the literature is explained by absence of the universal method or algorithm well suitable for all tasks of segmentation. Each method has its advantages and disadvantages. Therefore, in this paper had been set up the task of comparison the existing methods of detection and segmentation for MRT-images of brain tumors, comparing their positive and negative features. Authors had been considered, the examples of practical application of known segmentation methods from domestic and foreign sources such as threshold, region, edge detection, morphological watershed, clustering, based on atlases, using artificial neural networks. In article based on results of analysis has been shown that, the neural networks are able to solve the task of detection and segmentation of brain tumors on MRT-images automatically with high accuracy.
Key words: опухоли головного мозга, МРТ-изображения, распознавание образов, информационные технологии, решающие правила, методы сегментации, показатели качества сегментации, нейронные сети, brain tumors, MRT-images, pattern recognition, information technologies, dec