Original Article Proc IMechE Part H: J Engineering in Medicine 2020, Vol. 234(8) 794–811 Ó IMechE 2020 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/0954411920924496 journals.sagepub.com/home/pih Empirical mean curve decomposition with multiwavelet transformation for eye movements recognition using electrooculogram signals Harikrishna Mulam 1 and Malini Mudigonda 2 Abstract Many research works are in progress in classification of the eye movements using the electrooculography signals and employing them to control the human–computer interface systems. This article introduces a new model for recognizing various eye movements using electrooculography signals with the help of empirical mean curve decomposition and multi- wavelet transformation. Furthermore, this article also adopts a principal component analysis algorithm to reduce the dimension of electrooculography signals. Accordingly, the dimensionally reduced decomposed signal is provided to the neural network classifier for classifying the electrooculography signals, along with this, the weight of the neural network is fine-tuned with the assistance of the Levenberg–Marquardt algorithm. Finally, the proposed method is compared with the existing methods and it is observed that the proposed methodology gives the better performance in correspondence with accuracy, sensitivity, specificity, precision, false positive rate, false negative rate, negative predictive value, false dis- covery rate, F 1 score, and Mathews correlation coefficient. Keywords Electrooculography, signal, human–computer interface, neural network classifier, classification Date received: 5 March 2019; accepted: 17 October 2019 Introduction Nowadays, the importance of human–computer inter- face (HCI) 1 increases tremendously, which is highly concentrated on the communication between comput- ers and users. Moreover, the researchers are highly focussing on the technologies of HCI designing, which make communication in a more satisfiable and easy way. However, it is very important to process the elec- trical signals via the HCI system to establish a dialogue between humans and computers, for example, recogni- tion of the direction of eye movements, 1,2 classification of patients suffering from depression, 3 and controlling. Huge numbers of HCI systems are available to improve the ease and use of quality of interaction. Some popular HCI systems use signals from the human eye as control signals. It mainly worked based on eye tracking methods such as dual Purkinje image (DPI), infrared oculography (IROG), and search coils (SC). Moreover, many wearable electronic sensor systems are available to monitor human health, and it is possible by interfacing the computer electromyogram (EMG), electroencephalogram (EEG), electrocardiogram (ECG), polysomnogram (PSG), 4,5,6,7 monitor cardio- pulmonary activity, 8 and sense limb position 9 to assess the emotional and cognitive responses. Mostly, EEG and EMG systems have been investigated in various research topics to process and control computers and machines such as a wheelchair and robotic arm. 10–12 To decompose, the empirical mean curve decomposi- tion (EMCD) model is adopted. 13 Electrooculography (EOG) is used to record eye position and movements furnished by the electrical potential difference between two electrodes positioned on the skin on either side of the eye. Compared to other 1 Department of Electronics and Instrumentation Engineering, VNR Vignana Jyothi Institute of Engineering & Technology, Bachupally, Hyderabad, India 2 Department of Biomedical Engineering, University College of Engineering, Osmania University, Hyderabad, India Corresponding author: Harikrishna Mulam, Department of Electronics and Instrumentation Engineering, VNR Vignana Jyothi Institute of Engineering & Technology, Bachupally, Hyderabad-90, India. Email: harikrishna_m@vnrvjiet.in