SVM Based Classification of Neurodegenerative Diseases for Salient Brain Patterns
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2014, Vol 2, Issue 11
Abstract
The identification defects in the MRI brain images can save numerous lives. A method to implement the kernel function for feature extraction to identify the neurodegenerative Alzheimer disease in Brain Image is proposed. The input Brain image has converted into gray image and preprocessed and saliency map image is obtained from the preprocessed image. The saliency map obtained gives the intensity related information from the images. After getting the saliency map image we have to normalize the saliency map and applying the kernel fusion to the normalize image to extract the feature of the image. The normalization process refines the images pixels to certain extend and the exact information representing the different portions separately is obtained. The feature extraction process reduces the dimensionality of the image data to make the process more optimized and simple. Finally by using the SVM classifier is fed with features such as intensity, textural and statistical information, binary tissue segmentations or cortical thickness estimations. Overall proposed algorithm used to decrease the computational time and the presence of irrelevant and noisy features. The salient regions found with the proposed approach as systematically relevant for discrimination of AD patients this results completely coherent to what has been reported by clinical studies of AD.
Authors and Affiliations
S. Subashini, Mr. S. Sakkaravarthi
Keywords
Related Articles
- EP ID EP19020
- DOI -
- Views 310
- Downloads 9
Security in cloud over the virtual Environment
The Security issues in cloud concerns and mainly associated with security issues faced by cloud service providers and the service issues faced by the cloud customers. The administrator will track the information of both...
Data Mining Challenges With Big Data
Data analysis is a clear bottleneck in many applications, both due to lack of scalability of the underlying algorithms and due to the complexity of the data that needs to be analyzed. The value of data explodes when it...
Analysis On Hydraulic Performance of Kaplan Turbine Based on Elbow Draft Tube Geometry
This article focuses on the analysis on hydraulic performance of Kaplan Turbine based on Elbow draft tube geometry. A draft tube is a closed passage of gradually increasing cross sectional area which connects the runner...
Experimental Investigation and Material Formation of Aluminium With Graphite
the goal of this work is to automotive applications in terms of environmental, social, economic and technical aspects by establishing a set of quantifiable measures for design for sustainability that can be applied to....
An Efficient Technique for Fingerprint Features Protection and Person Identification Using Wavelet Transform
Fingerprint recognition is a widely used technique for person identification. The two major techniques for fingerprint identification are minutiae based technique and non-minutiae based technique. In this paper we propo...