Improved Segmentation of Neurodegenerative Disease using Deep Auto Encoders

Previous research has demonstrated that there is a correlation between a person body mass index (BMI) and their brain volume, and that correlation is typically an inverse one. However, the Body Mass Index (BMI), which does not take into account the fact that different areas of fat carry varying degrees of metabolic risk, is not necessarily the best indicator of health. This is because it does not consider the fact that different areas of fat carry different degrees of risk. In order to determine whether or whether there is a connection between MR-based fat depots and the quantity of gray matter (GM) in the brain, which is known to decrease in individuals, who suffer from neurodegenerative disorders. In this paper, Deep Auto Encoders (DAE) are used as a segmentation model to segment the neurodegenerative diseases in patients. Initially, the images are collected from the database; it is then pre-processed and segmented using DAE. The simulation is carried out in python to test the efficacy of unsupervised DAE segmentation over various conventional segmentation models. The results of simulation shows that the proposed method achieves higher rate of accuracy than other methods.