Deep learning methods to detect Alzheimer's disease from MRI: A systematic review

Alzheimer's disease (AD) is a progressive and irreversible neurodegenerative condition in the brain that affects memory, thinking, and behaviour. To overcome this problem, which according to the World Health Organization, is on the rise, creating strategies is essential to identify and predict the disease in its early stages before clinical manifestation. In addition to cognitive and mental tests, neuroimaging is promising in this field, especially in assessing brain matter loss. Therefore, computer-aided diagnosis systems have been imposed as fundamental tools to help imaging technicians as the diagnosis becomes less subjective and time-consuming. Thus, machine learning and deep learning (DL) techniques have come into play. In recent years, articles addressing the topic of Alzheimer's diagnosis through DL models are increasingly popular, with an exponential increase from year to year with increasingly higher accuracy values. However, the disease classification remains a challenging and progressing issue, not only in distinguishing between healthy controls and AD patients but mainly in differentiating intermediate stages such as mild cognitive impairment. Therefore, there is a need to develop more valuable and innovative techniques. This article presents an up-to-date systematic review of deep models to detect AD and its intermediate phase by evaluating magnetic resonance images. The DL models chosen by different authors are analysed, as well as their approaches regarding the used dataset and the data pre-processing and analysis techniques.

Automated summary

Alzheimer's disease is a progressive and irreversible neurodegenerative condition that affects memory, thinking, and behavior. Deep learning models show promise in diagnosing Alzheimer's using magnetic resonance images, but disease classification remains challenging. Developing more effective and innovative techniques is necessary.