Title :
Assessment of brain ventricular volumes from MRI images
Author :
Fabijanska, Anna
Author_Institution :
Inst. of Appl. Comput. Sci., Lodz Univ. of Technol., Lodz, Poland
Abstract :
This paper considers the problem of assessment of brain ventricular volumes from 3D MRI brain scans. It is an important problem because the enlargement of brain ventricles is a pathological condition which accompanies some brain diseases. However, the existing techniques of ventricular volumes measurement are still mainly manual or semi-manual. Therefore, this paper introduces methods for segmentation of the brain ventricular system and separation of the ventricles. In particular, the interactive segmentation approach based on the random walker segmentation algorithm is proposed. The introduced method firstly extracts the entire ventricular system of brain. Next, it separates the ventricles from each other. Finally, the method of assessment of ventricles volume is introduced. The results of applying the proposed approach to Brain Web MRI brain phantoms are presented and discussed. Both the visual and the numerical results compared to the anatomical data suggest the correctness of the introduced approach.
Keywords :
biomedical MRI; brain; diseases; image segmentation; medical image processing; random processes; 3D MRI brain scan; anatomical data; brain Web MRI brain phantom; brain disease; brain ventricular volume; interactive segmentation approach; pathological condition; random walker segmentation algorithm; ventricular volume measurement; Diseases; Fluids; Image segmentation; Magnetic resonance imaging; Three-dimensional displays; Visualization; Volume measurement; MRI; brain volume; image segmentation; random walk; ventricular system;
Conference_Titel :
Imaging Systems and Techniques (IST), 2014 IEEE International Conference on
Conference_Location :
Santorini
DOI :
10.1109/IST.2014.6958486