Image segmentation an early detection to Alzheimer's disease

Abstract

Image segmentation is a significant research field in medical engineering. Segmented brain images are used to visualize volume and quantitatively analyse anatomical and cortical structures. Segmented brain tissues provide an anatomical frame work for visualisation which has a potential use in neuro-science research and neurosurgical planning owing to advances in Magnetic Resonance Imaging (MRI). Segmentation refers to the labeling of pixels into different regions. At a higher level these regions can be labeled as anatomical structures and then grouped across different slices to give three dimensional descriptions of these structures. The method proposed here is successful in segmenting the brain without human intervention and accurately detecting the variation in the total volume or size of the brain. The purpose of this work is to detect any subtle changes in the volume of the brain.

Imaging and machine learning techniques for diagnosis of Alzheimer’s disease

“…Thus, extraction of the brain tissue and quantifying different types of issues is critical for the detection and diagnosis of AD. A good number of papers have been published on segmentation of brain tissues in AD (Mackay et al, 1996;Tanabe et al, 1997;Ashburner and Friston, 2000;Shattuck et al, 2001;Shanthi and Ravish, 2013).…”

“…This technique is error-prone and not very accurate especially in the presence of noise. Several works have reported separating brain tissue regions using this simple approach (Tanabe et al, 1997;Shanthi and Ravish, 2013). (2) Intensity histograms are created and used to define the threshold values (Smith, 2002;Nestor et al, 2013;Shanthi and Ravish, 2013).…”

“…(3) Adaptive thresholding algorithms are utilized to define an automated optimal threshold value. There are other thresholding techniques, such as global thresholding (Shanthi and Ravish, 2013) and iterative thresholding (Jentzen et al, 2007). A survey of different thresholding techniques and their performance comparison is provided by Sezgin and Sankur (2004).…”

“…Multiple thresholding is used in AD studies to separate different brain tissues to quantify the amounts of various tissue types (Tanabe et al, 1997;Shanthi and Ravish, 2013). Tanchi et al (2012) recommended values within the range T1 = 150 and T2 = 430 based on extensive observation to extract brain tissue regions in MRI images.…”

Segmentation of Brain Subjects in MR Images using Hybrid Segmentation Technique

“…Most of the Research demonstrates that these both proteins take place an important role in disease progress, but the exact reason for the AD is still not identified. [6] When the disease starts its progress, all the nerves and brain cells start damage over a time period. This damage of nerve cells leads a human to the beginning symptoms of Alzheimer's.…”