The programme was created and trained using EEG data from 88 human subjects. The group then categorised various wave patterns as sharp signals, spikes, and sluggish waves using data analysis.
An algorithm has been created by scientists from the Indian Institute of Science (IISc) and AIIMS Rishikesh that can analyse brain images to determine the presence and kind of epilepsy.
A neurological condition known as epilepsy causes the brain to suddenly and rapidly release large amounts of electrical signals, which can lead to seizures, fits, and in severe cases, even death.
Epilepsy is divided into two categories :
Focal Epilepsy and Generalised Epilepsy, depend on where in the brain the irregular impulses originate. Focal epilepsy is the medical term for irregular brain impulses that are confined to a specific region of the brain. If the signals come and go arbitrarily, it is referred to as generalised epilepsy.
EEGs are examined
Neurophysiologists must manually examine EEGs (electroencephalograms), which might record such irregular signals, to determine whether a patient is epileptic, according to a news statement from the IISc.
According to Hardik J. Pandya, assistant professor in the department of electronic systems engineering (DESE) and the corresponding author of the study that was published in biomedical signal processing and control, visual inspection of EEG can occasionally result in errors and grow tiring after repeated use.
In addition to attempting to distinguish between the EEGs of epileptic participants and normal people, the research aims to categorise the different types of seizures using an algorithm that was built.
The Algorithm’s Training
Each subject underwent a 45-minute EEG test, which was divided into two parts :
A 10-minute awake test, which involved photic stimulation and hyperventilation, and a 35-minute sleep session, during which the patient was told to nod off.
The wave patterns were separated into acute signals, spikes, and sluggish waves by the researchers after data analysis. According to the press release, spikes are patterns where a signal rises and falls quickly (in less than 70 milliseconds), sharps are patterns where the rises and falls are spaced out over a longer period (in more than 250 milliseconds), and slow waves have a much longer duration (in more than 400 milliseconds).
To determine whether an individual is epileptic or not, the researchers created an algorithm to count all sharp waves, known as the Cumulative Sharp Count.
Curves and spikes
The method computes the total of the regions below the spikes and sharp curves to distinguish between focal and generalised epilepsy, according to the press release (a higher value denotes generalised epilepsy, while a lower value denotes focal epilepsy).
According to the researchers, it demonstrates how to recognise absence seizures or those that include a brief, unexpected loss of consciousness; in some circumstances, these absence seizures are serious and may even be deadly.
The team then used their algorithm on a fresh batch of EEG data from individuals whose classification (whether they had epilepsy and, if so, what kind of epilepsy they had) was already known to the medical professionals. Nearly 91% of the time, this blind validation study correctly categorised the participants.
The algorithm is currently being tested for dependability by doctors at AIIMS Rishikesh, and a patent application has been submitted for the work.