Epileptic Classification With Deep-Transfer-Learning-Based Feature Fusion Algorithm

被引:34
作者
Cao, Jiuwen [1 ,2 ]
Hu, Dinghan [1 ]
Wang, Yaomin [1 ]
Wang, Jianzhong [1 ]
Lei, Baiying [3 ]
机构
[1] Hangzhou Dianzi Univ, Machine Learning & I Hlth Int Cooperat Base Zheji, Hangzhou 310018, Peoples R China
[2] Res Ctr Intelligent Sensing, Zhejiang Lab, Hangzhou 311100, Peoples R China
[3] Shenzhen Univ, Natl Reg Key Technol Engn Lab Med Ultrasound, Guangdong Key Lab Biomed Measurements & Ultrasoun, Sch Biomed Engn,Hlth Sci Ctr, Shenzhen 518060, Peoples R China
基金
中国国家自然科学基金;
关键词
Feature extraction; Electroencephalography; Databases; Brain modeling; Epilepsy; Transfer learning; Training; Deep transfer learning; deep neural networks (DNNs); mean amplitude spectrum (MAS); preictal classification; seizure detection; SEIZURES;
D O I
10.1109/TCDS.2021.3064228
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Epilepsy ictal detection based on scalp electroencephalograms (EEGs) has been comprehensively studied in the past decades. But few attentions have been paid to the preictal classification. In this article, a comprehensive study on epileptic state classification based on deep transfer learning (TL) is presented. The main contributions include: 1) the subband mean amplitude spectrum (MAS) map that characterizes the typical rhythms of brain activities is extracted for EEG representation; 2) five representative deep neural networks (DNNs) pretrained on ImageNet are applied for EEG feature TL; and 3) a 7-layer hierarchical neural network (HNN) that consists of three fully connected (Fc) and three dropout layers followed by a Softmax layer is developed to perform the epileptic state probability learning and classification. Experiments on the benchmark CHB-MIT and iNeuro EEG databases that contain several different types of seizures show that the proposed algorithm achieves the highest overall accuracies of 96.97% and 87.87% on the 5-state epileptic classification, respectively, that outperforms many existing state-of-the-art methods presented in this article.
引用
收藏
页码:684 / 695
页数:12
相关论文
共 58 条
[1]   Neuro-Detect: A Machine Learning-Based Fast and Accurate Seizure Detection System in the IoMT [J].
Abu Sayeed, Md ;
Mohanty, Saraju P. ;
Kougianos, Elias ;
Zaveri, Hitten P. .
IEEE TRANSACTIONS ON CONSUMER ELECTRONICS, 2019, 65 (03) :359-368
[2]   Deep convolutional neural network for the automated detection and diagnosis of seizure using EEG signals [J].
Acharya, U. Rajendra ;
Oh, Shu Lih ;
Hagiwara, Yuki ;
Tan, Jen Hong ;
Adeli, Hojjat .
COMPUTERS IN BIOLOGY AND MEDICINE, 2018, 100 :270-278
[3]  
Ahmedt-Aristizabal D, 2018, IEEE ENG MED BIO, P332, DOI 10.1109/EMBC.2018.8512249
[4]   Effect of fusing features from multiple DCNN architectures in image classification [J].
Akilan, Thangarajah ;
Wu, Qingming Jonathan ;
Zhang, Hui .
IET IMAGE PROCESSING, 2018, 12 (07) :1102-1110
[5]   Epileptic MEG Spike Detection Using Statistical Features and Genetic Programming with KNN [J].
Alotaiby, Turky N. ;
Alrshoud, Saud R. ;
Alshebeili, Saleh A. ;
Alhumaid, Majed H. ;
Alsabhan, Waleed M. .
JOURNAL OF HEALTHCARE ENGINEERING, 2017, 2017
[6]   Random forests [J].
Breiman, L .
MACHINE LEARNING, 2001, 45 (01) :5-32
[7]   Unsupervised Eye Blink Artifact Detection From EEG With Gaussian Mixture Model [J].
Cao, Jiuwen ;
Chen, Long ;
Hu, Dinghan ;
Dong, Fang ;
Jiang, Tiejia ;
Gao, Weidong ;
Gao, Feng .
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, 2021, 25 (08) :2895-2905
[8]   Maximum Correntropy Criterion-Based Hierarchical One-Class Classification [J].
Cao, Jiuwen ;
Dai, Haozhen ;
Lei, Baiying ;
Yin, Chun ;
Zeng, Huanqiang ;
Kummert, Anton .
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2021, 32 (08) :3748-3754
[9]   Epileptic Signal Classification With Deep EEG Features by Stacked CNNs [J].
Cao, Jiuwen ;
Zhu, Jiahua ;
Hu, Wenbin ;
Kummert, Anton .
IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS, 2020, 12 (04) :709-722
[10]   Extreme Learning Machine With Affine Transformation Inputs in an Activation Function [J].
Cao, Jiuwen ;
Zhang, Kai ;
Yong, Hongwei ;
Lai, Xiaoping ;
Chen, Badong ;
Lin, Zhiping .
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2019, 30 (07) :2093-2107