Abstract
Alzheimer's disease (AD) is one of many disorders that affect the brain. It is thought to have a number of neurological and physical causes, including the loss of muscle memory. It is characterized by an aggregation of clinical features; memory loss, confusion, and personality changes tend to worsen over time. The WHO stated that in 2021 there were more than 55 million people living with dementia, and this figure is estimated to increase to as high as 139 million by the year 2050. Alzheimer’s is responsible for 60-70% of dementia cases, with many other factors considered. Mild cognitive impairment (MCI) is a condition where individuals experience significant challenges in memory, cognition, and decision-making, often influenced by age and educational background. Individuals with MCI have two to three times the chance of advancing to Alzheimer's dementia compared to older adults, with an estimated annual transition rate of 3% to 15%. Various strategies across a range of AI tools, including machine learning and deep learning, are used for for AD diagnosis, particularly the application of machine learning algorithms. This paper presents the results of various research studies conducted in the last few years that particularly concern the path from MCI to AD. It focuses on reviewing machine learning and deep learning algorithms, covering convolutional neural networks and multitask learning techniques to predict Alzheimer's progression.