Multichannel EEG Pattern Acknowledgment Feature Assortment Using T-Test Ranking and Principal Component Analysis

Electroencephalography (EEG) is a non-invasive test used to assess human brain activity by utilizing signals generated from the brain; it is also used to detect epileptic seizures. Furthermore, this helps medical professionals identify conditions including epilepsy, seizure, sleeplessness, sleepiness, traumatic brain injury, headache, and brain tumor. It's also useful for establishing that a person has died of a brain injury. Epilepsy is a condition of the nervous system brought on by electrical spikes and waves. Epilepsy manifests itself in moments of amnesia, unconsciousness, and involuntary twitching. A person's behavior can have a significant impact on the activity of the brain's neurons. Some bodily functions are also affected by the shift in neuronal activity. Sensory, visual, and memory processing all occur in the brain in line with the flow of electrical signals. When the brain's nerve cells are compromised, a seizure results. Generalized seizures and focal seizures are the two main categories. This study proposes that for supervised learning tasks, it is possible to achieve better results using multichannel EEG recordings by selecting features with maximal class separability. In this study, we test the following theory: One definition of a good feature subset is one in which the features have a strong correlation with the class labels but no correlation with one another.