Non-neuroimaging can be as good as neuroimaging for the diagnosis and prognosis of AD
What is it about?
In this systematic review we found that the use of non-neuroimaging has increased in the last years for the diagnosis and prognosis of Alzheimer's Disease. The most used databases that include Alzheimer's Disease subjects has been indicated. Convolutional Neural Networks paired with neuroimaging techniques (Ex.: MRI and PET), the state-of-the-art, have superb performance for the diagnosis of Alzheimer's Disease, the most common type of dementia. However, using non-neuroimaging ones (Ex.: speech, gait, neuropsychological tests, genes, blood...) yields similar performance and, in most cases, does not need expensive, stationary or complex equipment.
Why is it important?
Unlike neuroimaging techniques, most non-neuroimaging ones can be used outside clinical settings and for a fraction of the cost. Considering their similar performance, both characteristics make non-neuroimaging useful for the study of Alzheimer's Disease, especially over time.