The Department of Information and Communication Engineering, Tongji University, Shanghai 201804, China.
Comput Intell Neurosci. 2020 Aug 1;2020:7691294. doi: 10.1155/2020/7691294. eCollection 2020.
In the research work of the brain-computer interface and the function of human brain work, the state classification of multitask state fMRI data is a problem. The fMRI signal of the human brain is a nonstationary signal with many noise effects and interference. Based on the commonly used nonstationary signal analysis method, Hilbert-Huang transform (HHT), we propose an improved circle-EMD algorithm to suppress the end effect. The algorithm can extract different intrinsic mode functions (IMFs), decompose the fMRI data to filter out low frequency and other redundant noise signals, and more accurately reflect the true characteristics of the original signal. For the filtered fMRI signal, we use three existing different machine learning methods: logistic regression (LR), support vector machine (SVM), and deep neural network (DNN) to achieve effective classification of different task states. The experiment compares the results of these machine learning methods and confirms that the deep neural network has the highest accuracy for task-state fMRI data classification and the effectiveness of the improved circle-EMD algorithm.
在脑机接口的研究工作和人脑功能研究中，多任务状态功能磁共振成像（fMRI）数据的状态分类是一个问题。人脑的 fMRI 信号是一种具有许多噪声影响和干扰的非平稳信号。基于常用的非平稳信号分析方法希尔伯特-黄变换（HHT），我们提出了一种改进的圆形经验模态分解（circle-EMD）算法来抑制端点效应。该算法可以提取不同的固有模态函数（IMFs），对 fMRI 数据进行分解以滤除低频和其他冗余噪声信号，更准确地反映原始信号的真实特征。对于过滤后的 fMRI 信号，我们使用三种现有的不同机器学习方法：逻辑回归（LR）、支持向量机（SVM）和深度神经网络（DNN）来实现不同任务状态的有效分类。实验比较了这些机器学习方法的结果，并证实深度神经网络对任务状态 fMRI 数据分类具有最高的准确性和改进的圆形 EMD 算法的有效性。
