The Evolution of Segmentation Techniques for Alzheimer's Disease Detection from Magnetic Resonance Images

Alzheimer's disease is a neurological sickness that is getting increasingly frequent and causes the structure and performance of the central nervous system to gradually deteriorate. The prevalence of Alzheimer's disease is expected to double by the year 2050. These illnesses pose a significant threat to the physical and mental well-being of individuals as well as their families across the globe. It is a sort of neurodegenerative sickness that is chronic and irreversible, and it is characterized by the gradual death of neurons. This death of neurons leads to reduced cognition and the loss of brain functions. There are presently 47 million Alzheimer's disease patients across the globe; however, it is expected that this number will more than double by the year 2050, attainment 152 million individuals. One of the several neuron imaging techniques that are currently at researchers' disposal is magnetic resonance imaging (MRI), which has made it possible for non-invasive in vivo studies of AD-related abnormalities in the brain. This research makes use of a number of different denoising techniques in order to cut down on noise and improve the overall picture quality. A clearer, more relevant and more easily digestible visual illustration is one of the goals that subdivision strives to achieve. For the purpose of segmenting the region of interest in MRI scans, a computer aided model has been presented. For the purpose of evaluating the suggested method of image segmentation, which has been validated with a large number of images, alternative image segmentation algorithms, including Otsu thresholding and region expanding, have been compared. It has been determined, using the performance evaluation metrics, that the proposed method is capable of achieving a higher level of accuracy than other methods.