Abstract: With the development of medical big data and artificial intelligence technology, the research on Alzheimer's disease (AD) based on structural magnetic resonance imaging (sMRI) using convolutional neural networks (CNN) has gradually become one of the hotspots in neuroscience. To further promote the application of 3D CNN in neuroimaging research, the research progress of sMRI analysis based on 3D CNN in AD classification was reviewed. First, the development of the machine learning technology applied to AD classification was reviewed. Then, the change of 3D CNN architecture and the research progress of its application in AD classification were introduced from the perspective of methods. Finally, the existing challenges and future development direction of the application of 3D CNN in the AD research field were discussed. It is expected that this technology can provide more accurate and effective help for early diagnosis of AD.