Detection of Alzheimer’s Disease Using Soft Computing Techniques

Abstract

Magnetic resonance imaging is non-invasive and is extensively used for evaluating the structures of
the brain and to diagnose the disease. Magnetic resonance imaging produce high quality and high
resolution images without exposing the human body into radiation. Magnetic resonance imaging is
preferred among various modalities in brain subject analysis since it has better visualization.
Alzheimer’s disease is a chronic condition which leads to behavioural and psychological problems.
The decisive and reliable segmentation in MR Imaging avail as an competent assist for the
physicians to diagnose the disorders at an earlier stage. There is no proven cure or method to
prevent alzheimer’s disease but there are few method to limit its progression. In this work, we
analyzed the segmentation of brain tissues to detect the Alzheimer's disease at early stage. The
existing method gives poor contrast and low computational time in alzheimer’s disease detection so
that it is very difficult to diagnose the disease for clinical usage. In this regard, we use fuzzy cmeans for improving quality parameters when compared to other algorithms like, random forest
algorithm. The design of clustering is to perform grouping of the similar kind of pixels together in a
group. Her, in this case we use the fuzzy c-means clustering accomplised for differentiating White
matter, Grey Matter, Cerebrospinal fluid and the computational time is also improved and therefore
this work is proved to be effective than other methods in clustering brain subjects.