Wavelet Transform-Statistical Time Features-Based Methodology for Epileptic Seizure Prediction Using Electrocardiogram Signals

Abstract

Epilepsy is a brain disorder that affects about 50 million persons around the world and is characterized by generating recurrent seizures, which can put patients in permanent because of falls, drowning, burns, and prolonged seizures that they can suffer. Hence, it is of vital importance to propose a methodology with the capability of predicting a seizure with several minutes before the onset, allowing that the patients take their precautions against injuries. In this regard, a methodology based on the wavelet packet transform (WPT), statistical time features (STFs), and a decision tree classifier (DTC) for predicting an epileptic seizure using electrocardiogram (ECG) signals is presented. Seventeen STFs were analyzed to measure changes in the properties of ECG signals and find characteristics capable of differentiating between healthy and 15 min prior to seizure signals. The effectiveness of the proposed methodology for predicting an epileptic event is demonstrated using a database of seven patients with 10 epileptic seizures, which was provided by the Massachusetts Institute of Technology–Beth Israel Hospital (MIT–BIH). The results show that the proposed methodology is capable of predicting an epileptic seizure 15 min before with an accuracy of 100%. Our results suggest that the use of STFs at frequency bands related to heart activity to find parameters for the prediction of epileptic seizures is suitable.

Method of Constructing a Nonlinear Approximating Scheme of a Complex Signal: Application Pattern Recognition

“…Currently, scientists are actively conducting research related to the development of methods for modeling and analyzing complex nonstationary signals [1][2][3]. The need to create such methods arises when carrying out a number of fundamental and applied investigations in such areas as biomedicine, geophysics, ecology, seismology, etc.…”

An Explainable Machine Learning Approach Based on Statistical Indexes and SVM for Stress Detection in Automobile Drivers Using Electromyographic Signals

“…STFs have demonstrated to be reliable tools for recognizing patterns or features in time signals with non-stationary properties, which have allowed for evaluating the health condition of structures [ 34 ] and induction motors [ 35 ], as well as predicting [ 33 ] and detecting diseases [ 36 , 37 ], among other applications. In general, STFs allow for measuring properties such as the range, asymmetry, convergence, and dispersion, among others, of time signals without needing to transform them into another domain [ 40 , 41 ].…”

“…In this regard, the Kruskal–Wallis method (KWM) was used for this step. This method can be used regardless of the probability distribution the database has [ 33 , 34 ], unlike the analysis of variance (ANOVA) method, which should be used with data that have a Gaussian (normal) distribution. Physiological data are known for having a non-Gaussian distribution [ 33 ]; hence, methods such as KWM are adequate to be used for this type of data.…”

“…KWM is a non-parametric method employed to estimate the statistical independence of datasets with non-normal distributions [ 45 , 46 ]. It has been employed for determining the most suitable features for detecting or estimating seizures [ 33 ], drowsiness levels [ 47 ], and respiration [ 48 ], among other applications. KWM evaluates the variation between datasets and the variations within each dataset [ 49 ] by calculating a probability value ( p -value).…”

“…In recent years, statistical time features (STFs) have demonstrated to be efficient for predicting seizures [ 33 ], evaluating the health condition of civil structures [ 34 ], induction motors [ 35 ], diagnosing neurodegenerative diseases [ 36 ], evaluating sleep disorders [ 37 ], and detecting alcohol use disorders [ 38 ], among other applications. They present great advantages such as low complexity, low computational costs, and are capable of measuring changes in non-stationary time signals [ 37 ], such as EMG signals, which can be useful for identifying relevant features into EMG signals in order to associate them to detect automobile drivers with stress.…”