基于实时功能磁共振成像的脑状态解码研究

近年来，基于功能磁共振成像(functional magnetic resonance imaging, fMRI)的脑状态解码研究成为认知神经科学的研究热点。当前有关解码研究可以分为三类：分类、识别和重构。本项目拟将从分类的角度来深入探讨基于实时fMRI(real-time fMRI或RT-fMRI)的脑状态实时解码和反馈。由于多变量模式分类方法被越来越多的用于fMRI解码研究中，本项目将针对模式分类方法在fMRI实时解码中存在的若干问题，围绕以下两个方面进行深入探讨：（1）基于RT-fMRI的脑状态解码的方法学研究。（2）基于RT-fMRI的情景记忆实时解码及反馈训练的研究。本项目的研究不仅能够更好的帮助我们去理解和改造大脑，同时也为我们开展更加灵活的认知能力训练实验以及临床的康复和治疗等方面提供一种可能的途径。

Recently, brain state decoding based on functional magnetic resonance imaging (fMRI) has become a hot topic of cognitive neuroscience. The types of decoding inlucde classification, identification and reconstruction. This project is going to deeply investigate brain state decoding and feedback based on real-time fMRI (RT-fMRI) technique from the classification. Because multivariate pattern classifcation is more and more widely applied to fMRI decoding studies, the current project will deeply investigate the following two critical issues in terms of the application of the pattern classification to the real-time decoding of fMRI data. 1) Methodological study of brain state decoding based on RT-fMRI. 2) Real-time decoding and feedback training of episodic memory based on RT-fMRI. This project will not only help us better understand and create our brain, but also provide a potential way for us to perform more flexible cognitive training and clinical rehabilitation.