A novel fuzzy pixel intensity correlation based segmentation algorithm for early detection of Alzheimer’s disease

Alzheimer’s disease (AD) is considered to be one of the most fatal neurological disorders and is identified as significant tissue loss in the hippocampus region of human brain. This paper presents a fuzzy based novel segmentation algorithm for brain MRI images. A structuring element for opening of g...

Full description

Saved in:
Bibliographic Details
Published inMultimedia tools and applications Vol. 78; no. 9; pp. 12465 - 12489
Main Authors Ghosh, Sukanta, Chandra, Abhijit, Mudi, Rajani K.
Format Journal Article
LanguageEnglish
Published New York Springer US 01.05.2019
Springer Nature B.V
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Alzheimer’s disease (AD) is considered to be one of the most fatal neurological disorders and is identified as significant tissue loss in the hippocampus region of human brain. This paper presents a fuzzy based novel segmentation algorithm for brain MRI images. A structuring element for opening of gray scale converted test MRI scans has been proposed in this regard. It effectively enhances the contrast of lateral ventricle region of brain which contains crucial information for mild cognitive impairment (MCI). Proposed rule-base of higher order fuzzy system dynamically chooses edge pixels and accordingly predicts the next probable edge pixel. Proposed fuzzy inference system is inspired by fuzzy connectedness algorithm and converts probable edge pixels into edge pixels depending on the intensity correlation between ordered pixels in support of rule-base and assembles it into an edge contour. Our proposition is finally tested over several ADNI brain images of different subject and orientation. Experimental results identify a promising improvement in detection of object boundaries and enhance contrast both qualitatively and quantitatively.
ISSN:1380-7501
1573-7721
DOI:10.1007/s11042-018-6773-z