The two-point central difference is a common algorithm in biological signal processing and is particularly useful in analyzing physiological signals. In this paper, we develop a model-based classification method to detect epileptic seizures that relies on this algorithm to filter EEG signals.The underlying idea is to design an EEG filter that enhances the waveform of epileptic signals. The filtered signal is fitted to a quadratic linear-parabolic model using the curve fitting technique. The model fitting is assessed using four statistical parameters, which are used as classification features with a random forest algorithm to discriminate seizure and non-seizure events. The proposed method was applied to 66 epochs from the Children Hospital Boston database. Results show that the method achieves fast and accurate detection of epileptic seizures, with a 92% sensitivity, 96% specificity, and 94.1% accuracy.Epilepsy is a neurological disease that affects people of all ages. It is characterized by unpredictable seizures resulting from the hyperexcitability of neurons. The electroencephalogram (EEG) is the predominant modality to study cerebral activity. The analysis of EEG signals is nearly completely dependent on visual inspection by the physician to quantify or qualify the morphology of waves, their goal being the identification and classification of abnormal patterns in order to provide aid for an epilepsy diagnosis.This study proposes a simple, fast and adaptable method, implementable in real-time, to help physicians visually inspect EEG signals for epileptic seizure detection. The proposed method fits within the framework of model-based classification. It is based on the statistical parameters obtained from fitting a quadratic parabolic model to specifically filtered and transformed EEG signals. Precisely, the two-point central difference algorithm is used to build a filter for the EEG signals. The filtered signal is subsequently represented using a quadratic linear-parabolic model, using the curve fitting approach. Four statistical parameters are then calculated to characterize the model fitting. These parameters are considered as a feature-vector in the classification stage.The literature abounds with research work dealing with the detection of epileptic seizure onset. Various methods and techniques have been proposed for this purpose. Many existing methods fit within the large framework of the feature-based machine learning approach.The two-point central difference algorithm has proven useful in analyzing physiological signals, with various medical applications reported in the literature. A typical work has been reported by Frei et al. to detect epileptiform discharges