Artificial neural networks for automatic ECG analysis

被引:161
作者
Silipo, R [1 ]
Marchesi, C
机构
[1] Univ Florence, Dipartimento Sistemi & Informat, Firenze, Italy
[2] CNR, Ist Fisiol Clin, Pisa, Italy
关键词
arrhythmia classification; ECG; ischemia detection; post-processing; pre-processing; recurrent neural networks; static neural networks;
D O I
10.1109/78.668803
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
The analysis of the ECG can benefit from the wide availability of computing technology as far as features and performances as well. This paper presents some results achieved by carrying out the classification tasks of a possible equipment integrating the most common features of the ECG analysis: arrhythmia, myocardial ischemia, chronic alterations. Several ANN architectures are implemented, tested, and compared with competing alternatives. Approach, structure, and learning algorithm of ANN were designed according to the features of each particular classification task, The trade-off between the time consuming training of ANN's and their performances is also explored. Data pre-and post-processing efforts on the system performance were critically tested. These efforts' crucial role on the reduction of the input space dimensions, on a more significant description of the input features, and on improving new or ambiguous event processing has been also documented. Finally the algorithm assessment was done on data coming from all the currently available ECG databases.
引用
收藏
页码:1417 / 1425
页数:9
相关论文
共 22 条
[1]  
BAILEY JJ, 1990, CIRCULATION, V81
[2]  
*BETH ISR HOSP BIO, KB26 BETH ISR HOSP B
[3]   LEARNING IN MULTILAYERED NETWORKS USED AS AUTOASSOCIATORS [J].
BIANCHINI, M ;
FRASCONI, P ;
GORI, M .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1995, 6 (02) :512-515
[4]  
Bortolan G., 1994, Journal of Electrocardiology, V26, P75
[5]   Fuzzy pattern classification and the connectionist approach [J].
Bortolan, G ;
Silipo, R ;
Marchesi, C .
PATTERN RECOGNITION LETTERS, 1996, 17 (06) :661-670
[6]  
BORTOLAN G, 1991, IEEE COMPUT CARDIOL, P177
[7]   AN APPROACH TO CARDIAC-ARRHYTHMIA ANALYSIS USING HIDDEN MARKOV-MODELS [J].
COAST, DA ;
STERN, RM ;
CANO, GG ;
BRILLER, SA .
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 1990, 37 (09) :826-836
[8]   NEURAL NETWORKS FOR CLASSIFICATION OF ECG ST-T SEGMENTS [J].
EDENBRANDT, L ;
DEVINE, B ;
MACFARLANE, PW .
JOURNAL OF ELECTROCARDIOLOGY, 1992, 25 (03) :167-173
[9]   UNIFIED INTEGRATION OF EXPLICIT KNOWLEDGE AND LEARNING BY EXAMPLE IN RECURRENT NETWORKS [J].
FRASCONI, P ;
GIRO, M ;
MAGGINI, M ;
SODA, G .
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 1995, 7 (02) :340-346
[10]  
Frenster J. H, 1990, P 12 ANN INT C IEEE, P1423