Envelope analysis and short time Fourier transform of EEG for epilepsy diagnosis enhancement

Background. Epilepsy is a chronic neurological disorder that affects millions worldwide, characterized by abnormal electrical activity in the brain. EEG (electroencephalogram) rhythms are critical for diagnosing epileptic seizures. This study uses EEG datasets from epilepsy patients recorded at the Neurology and Sleep Centre, Hauz Khas, New Delhi. Methods. A hybrid signal processing method combining envelope analysis and short-time Fourier transform (STFT) was proposed to enhance the diagnosis of epileptic seizures. The analysis compares EEG rhythms, envelope analysis, and STFT for three distinct stages: “preictal”, “interictal”, and “ictal”. This approach aims to identify seizure activity more effectively by examining changes in brain rhythms across these stages. Results. Envelope analysis showed a significant increase in amplitude during the ictal phase, with the amplitude in ictal cases being up to four times higher than in the interictal phase, reaching values greater than 200 µV. The time-frequency analysis using STFT illustrated the presence of lower frequencies in the ictal phase compared to the preictal and interictal phases, providing a clearer distinction between seizure states. Conclusion. The proposed method, combining envelope analysis with STFT, enhances epilepsy diagnosis by providing a more reliable and efficient assessment of EEG rhythms. Compared to conventional EEG-based diagnostic methods, this hybrid approach offers superior accuracy, with potential for improved real-time monitoring and evaluation of epilepsy.