Alzheimer’s disease, also known as AD, is a progressive neurodegenerative disorder leading to progressive cognitive decline that prevails in millions in the world. Timing is critical and testing is difficult because symptoms and other diseases may not be distinctive. Clinically reasoned techniques using evaluations may not be precise most of the time. Neuroimaging methods for PET and MRI together with multiple approaches have been identified to have features that can raise the degree of accuracy of diagnoses. However, issues such as minimal data sets or variation in imaging persist as issues to be solved. This review aims to present new trends in AD classification with an emphasis on the integrated use of imaging and computational algorithms for early and accurate diagnosis of Alzheimer’s disease.