INDIA (Commonwealth Union) – The electroencephalogram (EEG) is generally performed in the monitoring of neurodegenerative diseases such as Alzheimer’s disease, dementia together with tumors and also checking brain activity in relation to injuries and infections.
Researchers at the Indian Institute of Science (IISc), have joined hands with the All-India Institute of Medical Science Rishikesh (AIIMS), to produce an algorithm capable of decoding brain scans to mark the occurrence and type of epilepsy with EEGs.
Epilepsy is a neurological condition that usually brings about rapid shaking as a result of disruption to the brains electrical signals affecting millions worldwide. Finding out if a patient is epileptic, neurophysiologists have to manually inspect EEGs, which can identify erratic signals.
Visual observations of EEG can be laborious and time consuming, which can result in errors, according to Hardik J Pandya, Assistant Professor at the Department of Electronic Systems Engineering (DESE) and the corresponding author of the study that appeared in Biomedical Signal Processing and Control. “The research aims to differentiate EEG of normal subjects from epileptic EEGs. Additionally, the developed algorithm attempts to identify the types of seizures,” he said, adding that their work can assist neurologists make an efficient and speedy automated screening and diagnosis.
The researchers reported a novel algorithm that can screen EEG data and find signatures of epilepsy from the electrical signals with specific patterns. Following the 1st training, the algorithm was capable of detecting if a human subject could get epilepsy or not which was related to the patterns in their respective analyses with a highly accurate reading.
An epileptic subject indicates a unique set of patterns different to a healthy person. The researchers produced an algorithm to calculate the complete number of sharp waves, the increasing Sharp Count followed by the utilization of this as a parameter to identify if the subject is epileptic or not.
