CASPIAN JOURNAL

MANAGEMENT AND HIGH TECHNOLOGIES

Automatic detection of acute stroke zone on MRI

Read Dmitriev Gennadiy A., Kirsanova Anna V., Al-Bakheli Vasel Akhmed Akhmed  Automatic detection of acute stroke zone on MRI // Caspian journal : management and high technologies. — 2014. — №4. — pp. 166-174.

Dmitriev Gennadiy A. - D.Sc. (Engineering), Professor, Honored Scientist of Russia, Tver State Technical University, 22 Afanasiy Nikitin St., Tver, 170026, Russian Federation, dmitrievtstu@mail.ru

Kirsanova Anna V. - Ph.D. (Engineering), Associate Professor, Tver State Technical University, 22 Afanasiy Nikitin St., Tver, 170026, Russian Federation, kirsanova_av@mail.ru

Al-Bakheli Vasel Akhmed Akhmed  - postgraduate student, Tver State Technical University, 22 Afanasiy Nikitin St., Tver, 170026, Russian Federation, wasel2800@mail.ru

The article presents an algorithm for the automatic detection of the acute stroke zone on the MRI. To searching for a zone of interest used a combination of the tomograms of the brain of various types. After determining the coordinates and boundaries of the pathological signal field on the image obtained by subtracting images T1 and Flair, these coordinates are automatically transferred to the diffusion-weighted images of type b-1000, obtained from the same patient with the same isocenter coordinates. Next, calculated the brightness of the selected area on the diffusion-weighted images. If the brightness value more than a certain threshold value, it is concluded that the abnormal region is an acute ischemic stroke. If the brightness is below a threshold value, it is said from the fact that the region is not an acute ischemic stroke

Key words: автоматическая сегментация, распознавание магнитно-резонансных изображений, автоматизация диагностики ишемического инсульта, системы поддержки принятия решений, пороговые методы обработки изображений, морфологические методы обработки изображений, Automati