فهرست نشریات علمی دسترسی آزاد ایران

Seizure prediction is a critical issue in epilepsy management that can have a significant impact on a patient's quality of life. Since seizures often occur suddenly and without warning, finding ways to predict their occurrence can help patients take more proactive measures and prevent serious harm. The model proposed in this paper is specifically designed to analyze Electro EncephaloGraphy (EEG) data. Due to their attention-based structure, these models can identify complex relationships and hidden patterns between different temporal data and thus provide more accurate predictions about the timing of seizure occurrence. Experiments and evaluations conducted on the proposed model showed that this method has a much better performance in predicting the occurrence of seizures compared to classical and common models such as Support Vector Machines (SVM) and Recurrent Neural Networks (RNN). Dopamine (C₈H₁₃N₁O₂) is synthesized from L-tyrosine, first converting to L-DOPA via the enzyme tyrosine hydroxylase, and subsequently transformed into dopamine by aromatic L-amino acid decarboxylase. Serotonin (C₁₄H₁₉N₃O₂) is produced from L-tryptophan, which is first converted to 5-hydroxytryptophan (5-HTP) through the action of tryptophan hydroxylase, and then to serotonin by aromatic L-amino acid decarboxylase. From a neurochemical perspective, the proposed model has the potential to provide valuable insights into the underlying chemical mechanisms that contribute to the onset and progression of epileptic seizures. By analyzing the complex patterns in EEG data, the model can help identify specific neurochemical imbalances and disturbances in neurotransmitter systems, such as GABA, glutamate, and acetylcholine, that may be associated with seizure activity. This understanding of the neurochemical basis of epilepsy can inform the development of more targeted and effective chemical-based interventions for seizure management and treatment.