Title :
Cerebral abnormalities detection by region-growing segmentation and KNN classification
Author :
Hachemi, B. ; Oudjemia, S. ; Alim, F. ; Seddiki, S. ; Talbi, F. ; Abdelaziz, M.
Author_Institution :
Div. of Syst. Archit. & Multimedia (ASM), Centre for Dev. of Adv. Technol. (CDTA), Algiers, Algeria
Abstract :
In this paper, we propose a computer aided diagnosis system for brain MRI. This system is based on region-growing segmentation and KNN classification. The co-occurrence matrix and the discrete wavelet decomposition methods are used to extract the parameters of suspect region. At the last part, we applied our system to other subject to estimate its accuracy.
Keywords :
biomedical MRI; discrete wavelet transforms; feature extraction; image classification; image segmentation; matrix algebra; medical image processing; KNN classification; brain MRI; cerebral abnormality detection; co-occurrence matrix; computer aided diagnosis system; discrete wavelet decomposition methods; region-growing segmentation; suspect region parameter extraction; Computers; Image segmentation; Magnetic resonance imaging; Matrix decomposition; Medical diagnostic imaging; Tumors; Diagnostic support; KNN; MRI brain; Region-Growing; Tumor; segmentation;
Conference_Titel :
Innovative Computing Technology (INTECH), 2015 Fifth International Conference on
Conference_Location :
Galcia
DOI :
10.1109/INTECH.2015.7173371
