Nitric oxide (NO) is known to be a neuromodulator with dual proconvulsive and anticonvul- sive action. Valeriana officinalis (VAL) was previously believed to be antiepileptic, but is today known as a sedative and sleep regulator. Seizures may be associated with abnormal electrocardio- graphic changes and cardiac dysfunction arising from epilepsy may be related with neuronal nitric oxide (nNO). This study was aimed to investigate the effects of the neuronal nitric oxide synthase (nNOS) inhibitor 7-Nitroindazole (7-NI) and VAL on seizure behaviours and electrocar- diographic parameters in the pentylentetrazole (PTZ)-kindled seizure model.
Wistar rats were randomised into saline control, PTZ-kindled, 7-NI, VAL and VAL+PTZ, 7-NI+PTZ and VAL+7-NI+PTZ groups. Latency, stage, frequency of seizures, blood pressure (BP), heart rate (HR) and corrected QT (QTc) values were evaluated.
Frequency and stage of seizures, BP and HR increased, while seizure latency decreased and QTc was prolonged in the PTZ-kindled group. 7-NI and VAL had no effects on BP and HR variables under normal conditions, but ameliorated the seizure stage and frequency of seizures. 7-NI treatment also resulted in a reduction of the increased BP and prolonged QTc values observed in PTZ-kindled rats.
Considering these results, QTc prolongation may be used as a predictor for recurrent seizures. 7-NI and VAL exhibited different effects on seizures and ECG variables. 7-NI shows potential as an anticonvulsant drug agent in epileptic patients with cardiac dysfunctions and those additional studies including in-vivo experiments are essential.
An application specific integrated design using Quadrature Linear Discriminant Analysis is proposed for automatic detection of normal and epilepsy seizure signals from EEG recordings in epilepsy patients. Five statistical parameters are extracted to form the feature vector for training of the classifier. The statistical parameters are Standardised Moment, Co-efficient of Variance, Range, Root Mean Square Value and Energy. The Intellectual Property Core performs the process of filtering, segmentation, extraction of statistical features and classification of epilepsy seizure and normal signals. The design is implemented in Zynq 7000 Zc706 SoC with average accuracy of 99%, Specificity of 100%, F1 score of 0.99, Sensitivity of 98% and Precision of 100 % with error rate of 0.0013/hr., which is approximately zero false detection.