Computer Methods and Programs in Biomedicine 151 (2017) 71–78 Contents lists available at ScienceDirect Computer Methods and Programs in Biomedicine journal homepage: www.elsevier.com/locate/cmpb Fusion of ECG and ABP signals based on wavelet transform for cardiac arrhythmias classification Roghayyeh Arvanaghi a , Sabalan Daneshvar a,b,∗ , Hadi Seyedarabi a,b , Atefeh Goshvarpour c a Department of Biomedical Engineering, Faculty of Advanced Medical Science, Tabriz University of Medical Sciences, Tabriz, Iran b Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran c Department of Biomedical Engineering, Faculty of Electrical Engineering, Sahand University of Technology, Tabriz, Iran a r t i c l e i n f o Article history: Received 30 December 2016 Revised 29 July 2017 Accepted 21 August 2017 Keywords: Atrial Blood Pressure (ABP) Discrete Wavelet Transformation (DWT) Electrocardiogram (ECG) Fusion Multi-Layer Perceptron Neural Network (MLPNN) a b s t r a c t Background and Objective: Each of Electrocardiogram (ECG) and Atrial Blood Pressure (ABP) signals con- tain information of cardiac status. This information can be used for diagnosis and monitoring of diseases. The majority of previously proposed methods rely only on ECG signal to classify heart rhythms. In this paper, ECG and ABP were used to classify five different types of heart rhythms. To this end, two men- tioned signals (ECG and ABP) have been fused. Methods: These physiological signals have been used from MINIC physioNet database. ECG and ABP sig- nals have been fused together on the basis of the proposed Discrete Wavelet Transformation fusion tech- nique. Then, some frequency features were extracted from the fused signal. To classify the different types of cardiac arrhythmias, these features were given to a multi-layer perceptron neural network. Results: In this study, the best results for the proposed fusion algorithm were obtained. In this case, the accuracy rates of 96.6%, 96.9%, 95.6% and 93.9% were achieved for two, three, four and five classes, respectively. However, the maximum classification rate of 89% was obtained for two classes on the basis of ECG features. Conclusions: It has been found that the higher accuracy rates were acquired by using the proposed fusion technique. The results confirmed the importance of fusing features from different physiological signals to gain more accurate assessments. © 2017 Elsevier B.V. All rights reserved. 1. Introduction Biosignals are nonstationary in nature; therefore, they may oc- cur accidentally through the time. For example, the symptoms are observed irregularly or may not be shown during the day at all. Hence, it is necessary to monitor a patient for a long time which causes the data complexity, the physician should have more time to interpret the data. Moreover, the human observer cannot di- rectly monitor all details from the signal; Thus, the computer- based analysis and the classification can help the physician to diag- nose accurately [1,2]. On the other hand, the main problem in ECG analysis is the large morphological variation of ECG waveforms, not only in different patients or patient groups but also within the same patient [3]. ∗ Corresponding author at: Faculty of Electrical and Computer Engineering, Uni- versity of Tabriz, Tabriz, Iran. E-mail addresses: r.arvanaghi88@gmail.com (R. Arvanaghi), daneshvar@tabrizu.ac.ir (S. Daneshvar), seyedarabi@tabrizu.ac.ir (H. Seyedarabi), af_goshvarpour@sut.ac.ir (A. Goshvarpour). Additionally, it is difficult to extract comprehensive information from the ECG signal itself in the assessment of heart health. It seems that involving other biosignals arising from the cardiac ac- tivity will increase the accuracy of diagnosis [2,4]. To extract useful information from the biological signal fluctuations, it is important to apply robust and appropriate processing techniques. Recently, some researchers are encouraged to use data fusion techniques to improve the results [5]. Data fusion algorithms were applied to combine information (data) acquired from several sources, information processing blocks or from one source through the time [6]. In other words, data fu- sion is the process of integration of multiple data and knowledge representing the same real-world object into a consistent, accurate, and useful representation [7]. Recently, some researchers were encouraged to apply fusion techniques on the ECG and other biological signals for different purposes. Ding et al. [4] fused ECG signal with blood pressure to detect the QRS peaks more accurately. For QRS peak detection, their proposed fusion technique outperformed other investigations (that have been relied only on ECG signal); especially for discover- ing the peaks of missing data or noisy ECG signal. This robustness http://dx.doi.org/10.1016/j.cmpb.2017.08.013 0169-2607/© 2017 Elsevier B.V. All rights reserved.