深度学习(DL)通过启用由多个处理层组成的计算模型来学习数据的抽象表示,从而彻底改变了人工智能的领域(Hinton等,2006; Bengio等,2006)。传统的机器学习方法数十年来一直限制,因为需要专家知识来设计复杂的特征提取算法,这在将原始数据转换为合适的分类形式的过程中。相比之下,深层倾斜的方法作为表示学习技术,使学习模型能够直接用原始数据馈送,以发现分类所需的表示形式(Krizhevsky等,2017; Lecun等,2015)。Currently, an intensive research effort is being devoted to the development of novel neuroimaging techniques to better understand the mechanisms of the central nervous system (CNS) and to early recognize age-related neural diseases ( Payan and Montana, 2015; Sarraf and Tofighi, 2016; Martinez-Murcia et al., 2020; Martinez-Murcia et al., 2018, 2016 ) Ortiz等。。大量多中心研究提供的大量数据调查了与年龄相关的神经疾病的新生物标志物,这为开发更准确的深度学习模型提供了一个机会,以早期认识神经退行性变化以及神经疾病的渐进过程(Cole and Franke,2017; Marzban et et and e an e an Al an Al a al niz and an an an an e an e e e an and and and and and and and and and an e e e e e e e e e e e e e e e e eT an and and and and and and and and and and and and and and and and。等,2018)。
主要关键词