The rapidly growing field of electroencephalography (EEG)-driven Seizure Detection Systems (SDSs) has attracted significant attention in the healthcare industry, focusing mainly on creating innovative methods for the early detection of epileptic seizures. Epilepsy, a neurological disorder marked by recurrent seizures, results from sudden changes in brain electrical activity. A traditional electroencephalogram (EEG) records the synchronized electrical impulses produced by the brain. Building on this principle, a new IoT-enabled EEG system is proposed to monitor and analyze multichannel EEG data. The system includes two key components: the multichannel EEG recording module and the seizure detection module. The main goal of this research is to design and develop an optimized seizure detection module that employs the Flower Pollination Algorithm (FPA) along with a CNN classifier within an IoT-supported EEG monitoring system to detect seizures. This proposed system offers improved performance compared to previous algorithms, with significant gains in accuracy using the CNN classifier. The effectiveness of this approach is expected to enhance the analysis of seizure data, especially in wearable medical devices.