Sonification is a New Step Forward in Diagnosing Epilepsy

Diagnosing epilepsy has taken a step forward with the development of a new auditory display technique known as sonification.
 
Sonification is an auditory display technique for representing a sequence of data values as sound says Doctor Alistair McEwan,  who co-ordinated the research team at The University of Sydney.
 
“An EEG records and measures the electrical activity of the brain.  The key brain wave signals associated with epilepsy repeat about five times per second.   But this frequency is too low for the human ear to hear so using sonification we speed up the signal by 60 times. At that speed the normal brain activity becomes audible and sounds like normal background noise, for example, a 'murmur of voices' and a squeaky computer or air conditioning fan.”
 
“Seizures are easily identified as they are associated with a rapid increase in the pitch,” says Doctor McEwan.
There are up to 800 thousand Australians living with epilepsy.
Because this audio detection method only requires a few hours of training, it offers an exciting possibility for a person living with epilepsy or their carer to collect information about their condition. Working in conjunction with the medical specialists, this information would be valuable for assessment and determination of medication.
The sonification team’s method has been tested on a group of non-experts.  Doctor Heba Khamis says: “What is great is that participants in our study at the University of Sydney’s Faculty of Engineering and Information technologies spent two hours in a training session where they learned how to audibly distinguish between seizures and some common sounds. “We found the participants’ accuracy in audio detection was very similar to the accuracy of visual detection. “ And training for visual detection requires a full year of training,” says Doctor Khamis.
The researchers hope to take their pioneering research to the next phase of clinical trials and develop a portable EEG system.
Doctor McEwan received support from Microsoft Research to commence this project through its Faculty Fellowship Program