Computer aided tool for Early Detection of Alzheimer’s Disease

Abstract

Over recent years, medical imaging research has increased to its zenith that employs principles of engineering utilizing Artificial Intelligence techniques and image computing techniques to medical modalities to yield efficient healthcare solutions. Alzheimer’s disease (AD) is a devastating condition that leads to significant memory loss due to neurodegeneration. Early identification of such conditions is significant to reduce the disease progression which subsequently improves the life of patients. MRI scan has been the best modality to visually diagnose the brain tissue state. Visual inspection by the radiologists has its limitations to accurately detect the amount of cell death happened. Moreover, such a volumetric analysis is of great importance to identify the progression rate of neurodegeneration of brain tissues over a period of time. Hence an automated system that provides a volume of neuronal death of individual brain tissues from MRI scan images helps to reduce time and effort of radiologists to diagnose the disease progression. This research aimed to develop  a system that accurately measure the pixel volume of brain tissues. The methodology included  image pre-processing steps followed by segmentation of brain tissues into Gray matter, White matter and Cerebrospinal fluid. The outcome of research is an application that automatically segments the brain tissues and obtain the pixel volume which has been successfully developed using digital image computing and unsupervised machine learning under Artificial Intelligence algorithms. The application helps the radiologists to input the MRI brain images and get the volume of each tissue from the inputted MRI DICOM brain images.