Near real-time single-beat myocardial infarction detection from single-lead electrocardiogram using Long Short-Term Memory Neural Network

被引:14
作者
Martin, Harold [1 ]
Izquierdo, Walter [1 ]
Cabrerizo, Mercedes [1 ]
Cabrera, Anastasio [2 ]
Adjouadi, Malek [1 ]
机构
[1] Florida Int Univ, Dept Elect & Comp Engn, CATE, Miami, FL 33199 USA
[2] Manatee Mem Hosp, Bradenton, FL USA
基金
美国国家科学基金会;
关键词
Cardiovascular disease; Electrocardiograms; Long Short-Term Memory Neural Network; Myocardial infarction; Real-time processing; 12-LEAD ECG; CLASSIFICATION; IDENTIFICATION; SIGNALS; COMPLEX; ENERGY;
D O I
10.1016/j.bspc.2021.102683
中图分类号
R318 [生物医学工程];
学科分类号
0831 ;
摘要
This study proposes a novel Long Short-Term Memory Neural Network (LSTM) architecture for the diagnosis of myocardial infarctions from individual heartbeats of single-lead electrocardiograms (ECGs). The proposed model is trained using an unbiased patient split approach and validated using 10-fold cross-validation over 148 myocardial infarction and 52 Healthy Control patients from the Physikalisch-Technische Bundesanstalt diagnostic ECG Database to generate an inter-patient classifier. We further demonstrate why special care must be taken when generating the training and testing datasets by exploring the effects of various data-split techniques that could mask the occurrence of overfitting and produce misleadingly high testing metrics of the model's performance. A thorough assessment of these results is provided using several standard metrics for different data split methods to show their tendency to overfitting, data leakage, and bias introduced from previously seen heart beats during the training phase. The design achieves near real-time diagnosis of 40 ms while providing an accuracy of 89.56% (with a 95% Confidence Interval (CI) of +/- 2.79%), recall/sensitivity of 91.88% (+/- 3.13% 95% CI), and a specificity of 80.81% (+/- 9.62% 95%CI). The fast processing makes the model readily deployable on currently existing mobile devices and testing instruments. The achieved performance makes the proposed method a new research direction for attaining real-time and unbiased diagnosis. While, the modular architectural design of the LSTM network structure, which is amenable for the inclusion of other ECG leads, could serve as a platform for early detection of myocardial infarction and for the planning of early treatment(s).
引用
收藏
页数:11
相关论文
共 57 条
[1]   Application of deep convolutional neural network for automated detection of myocardial infarction using ECG signals [J].
Acharya, U. Rajendra ;
Fujita, Hamido ;
Oh, Shu Lih ;
Hagiwara, Yuki ;
Tan, Jen Hong ;
Adam, Muhammad .
INFORMATION SCIENCES, 2017, 415 :190-198
[2]   Automated characterization of coronary artery disease, myocardial infarction, and congestive heart failure using contourlet and shearlet transforms of electrocardiogram signal [J].
Acharya, U. Rajendra ;
Fujita, Hamido ;
Sudarshan, Vidya K. ;
Oh, Shu Lih ;
Adam, Muhammad ;
Tan, Jen Hong ;
Koo, Jie Hui ;
Jain, Arihant ;
Lim, Choo Min ;
Chua, Kuang Chua .
KNOWLEDGE-BASED SYSTEMS, 2017, 132 :156-166
[3]   Automated detection and localization of myocardial infarction using electrocardiogram: a comparative study of different leads [J].
Acharya, U. Rajendra ;
Fujita, Hamido ;
Sudarshan, Vidya K. ;
Oh, Shu Lih ;
Adam, Muhammad ;
Koh, Joel E. W. ;
Tan, Jen Hong ;
Ghista, Dhanjoo N. ;
Martis, Roshan Joy ;
Chua, Chua K. ;
Poo, Chua Kok ;
Tan, Ru San .
KNOWLEDGE-BASED SYSTEMS, 2016, 99 :146-156
[4]   Machine learning-based coronary artery disease diagnosis: A comprehensive review [J].
Alizadehsani, Roohallah ;
Abdar, Moloud ;
Roshanzamir, Mohamad ;
Khosravi, Abbas ;
Kebria, Parham M. ;
Khozeimeh, Fahime ;
Nahavandi, Saeid ;
Sarrafzadegan, Nizal ;
Acharya, U. Rajendra .
COMPUTERS IN BIOLOGY AND MEDICINE, 2019, 111
[5]  
[Anonymous], 2017, HEART ATTACK
[6]  
[Anonymous], 2017, Heart Disease Fact Sheet|Data Statistics|DHDSP|CDC
[7]  
Bansal K, ANTERIOR MYOCARDIAL
[8]   Case histories Acute myocardial infarction [J].
Barnett, Richard .
LANCET, 2019, 393 (10191) :2580-2580
[9]   APPLICATION OF ARTIFICIAL NEURAL NETWORKS TO CLINICAL MEDICINE [J].
BAXT, WG .
LANCET, 1995, 346 (8983) :1135-1138
[10]   A neural computational aid to the diagnosis of acute myocardial infarction [J].
Baxt, WG ;
Shofer, FS ;
Sites, FD ;
Hollander, JE .
ANNALS OF EMERGENCY MEDICINE, 2002, 39 (04) :366-373