Title :
Automatic brain MR images diagnosis based on edge fractal dimension and spectral energy signature
Author :
Lahmiri, Salim ; Boukadoum, Mounir
Author_Institution :
Dept. of Comput. Sci., Univ. of Quebec at Montreal, Montreal, QC, Canada
fDate :
Aug. 28 2012-Sept. 1 2012
Abstract :
A new automatic system to detect pathologies in human brain magnetic resonance (MR) images is presented. The goal is to classify normal versus abnormal images affected by Alzheimer, Glioma, Herpes, Metastatic, and Multiple Sclerosis. The extracted features are the fractal dimension of edges in the Hilbert domain, and the skewness and kurtosis of their spectral energy distribution. The proposed system (FDSE) outperforms the popular discrete wavelet transform (DWT) and principal component analysis (PCA).
Keywords :
Hilbert transforms; biomedical MRI; brain; diseases; edge detection; feature extraction; image classification; medical image processing; Alzheimer disease; FDSE; Hilbert domain; automatic brain MR image diagnosis; automatic system; edge fractal dimension; feature extraction; glioma; herpes; human brain magnetic resonance image; metastasis; multiple sclerosis; pathology detection; spectral energy distribution kurtosis; spectral energy distribution skewness; spectral energy signature; Discrete wavelet transforms; Feature extraction; Fractals; Image edge detection; Pathology; Principal component analysis; Automation; Brain Diseases; Fractals; Humans; Magnetic Resonance Imaging;
Conference_Titel :
Engineering in Medicine and Biology Society (EMBC), 2012 Annual International Conference of the IEEE
Conference_Location :
San Diego, CA
Print_ISBN :
978-1-4244-4119-8
Electronic_ISBN :
1557-170X
DOI :
10.1109/EMBC.2012.6347421
