Abstract: Alzheimer’s disease is a common neurodegenerative disease, which can be clinically diagnosed according to neuroimaging. Deep learning can mine the rich information hidden in the patient’s image data and complete the classification of the course of disease in different stages, which is the research hotspot in the field of computer aided diagnosis. This paper presents the Alzheimer’s disease neuroimaging data set. This paper summarizes the application of the classical deep learning network model in the classification and diagnosis of Alzheimer’s disease and the interpretability of the deep learning model, and focuses on the classification and diagnosis methods of convolutional neural networks and multi-network convergence, then the author makes a comprehensive comparison of different ideas and methods. The article discusses the challenges and future research directions of deep learning in the field of Alzheimer’s disease auxiliary diagnosis, which is of great significance to improve the clinical diagnosis efficiency and early prediction accuracy of Alzheimer’s disease.

Key words: Alzheimer’s disease, deep learning, computer aided diagnosis, convolutional neural network

摘要： 阿尔茨海默病是一种常见的神经退行性疾病，可依据神经影像学进行临床诊断。深度学习能够挖掘患者影像资料中隐含的丰富信息并完成不同阶段的病程分类，是目前计算机辅助诊断领域的研究热点。介绍阿尔茨海默病神经影像学数据集，总结经典深度学习网络模型在阿尔茨海默病分类诊断中的应用以及深度学习模型可解释性，重点对卷积神经网络与融合多网络的分类诊断方法进行梳理分析，对不同的思路和方法综合对比，讨论深度学习在阿尔茨海默病辅助诊断领域面临的挑战与未来研究方向，对提高阿尔茨海默病的临床诊断效率与早期预测准确性具有重要意义。

关键词: 阿尔茨海默病, 深度学习, 计算机辅助诊断, 卷积神经网络