Premorbid brain volume against Alzheimer's disease using multimodal big medical data analysis

Abstract

Abstract Alzheimer’s disease is a neurological ailment in which memory loss and cognitive impairment are brought on by the death of brain cells. Alzheimer's disease is the most prevalent kind of dementia affecting people aged 60 and above. It is a neurodegenerative type of dementia that starts in the middle and gets worse over time. Alzheimer's disease, epilepsy, multiple sclerosis, cancer, depression, and other brain illnesses can all be diagnosed using hippocampus segmentation. Medical pictures have had a significant impact on medicine, diagnosis, and treatment. One of the most crucial image processing techniques is called image segmentation. Our research focuses on measuring the volume concerning typical size by utilizing segmentation techniques. To comprehend the severity of progression in demented people, this study will look at the whole brain (WB), grey matter (GM), and hippocampal (HC) morphological variation and identify the significant biomarkers in MRI brain images. Pre-trained models can demonstrate hippocampal regions with significant severity differences for the considered classes of CN and AD. It is determined that the CNN model for the HC region produces better categorization for CN and AD with 98.2 percent accuracy each. The primary goal of the research was to identify size anomalies using several biochemical features of big medical data analysis.