A Simple and Robust Strategy for Background Removal from Brain MR Images

International Conference on Recent Advancements and Effectual Researches in Engineering Science and Technology

Rameesa Mol, J.^1 ; Rashmi, A.G.^1 ; Jiss Mohan, K.^1 ; Alsaheer, S.S.^1 ; Mable, A.^1 ; Justin, J.^1

School of Electronics, St Joseph's College of Engineering and Technology, Pala, Kerala, India^1

关键词: Background removal;    Brain magnetic resonance images;    Contrast Enhancement;    Gradient based thresholding;    Morphological operations;    Morphological structures;    Qualitative evaluations;    Regions of interest;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/396/1/012039/pdf
DOI  :  10.1088/1757-899X/396/1/012039

来源: IOP

【 摘 要 】

This paper proposes a simple and robust method for elimination of noisy background region from the brain Magnetic Resonance Images. Background beneath the morphological structures presents strong pixel intensity inhomogeneity. This pixel intensity inhomogeneity may get preserved during non-linear edge sensitive smoothening and further may get enhanced during contrast enhancement. This inhomogeneous and noisy background may further hamper the accuracy of skull stripping and segmentation of morphological regions of interest. The skull stripping becomes easier if background could be eliminated prior to it. Background elimination is accomplished by multiplying the raw MR image with a multiplication mask synthesized by gradient based thresholding and a series of morphological operations. Qualitative evaluation of the results confirms the efficiency of the proposed method.