Seizure detection using multiple biosignals more accurate than detection by heart rate changes alone

According to data presented at the 2015 American Epilepsy Society Annual Meeting in Philadelphia seizure detection using multiple extracerebral biosignals specifically [HR], [SpO2] and electrodermal activity [EDA] results in improved accuracy compared with heart rate alone.

Using 2 commercially available wrist-worn devices, Cogan and colleagues gathered HR, arterial oxygenation (SpO2), and EDA data from 20 patients electively admitted to an epilepsy-monitoring unit. Of these patients, 11 provided HR, SpO2, and EDA data on 24 seizures during 355 hours of data collection.

In all 24 of the captured seizures HR increased by 15% or more. In 20 of these seizures, an SpO2 drop of 5% or more was seen immediately following the HR rise. This combination of HR↑ + SpO2↓ is less likely to be seen in non-seizure situations than an increase in HR alone.

Two complex partial seizures (CPS) resulted in less pronounced SpO2 disturbances which might require capture with more advanced (e.g. wavelet based) signal processing. SpO2 was undisturbed in only two CPS.

Of the 20 seizures with HR↑ + SpO2↓ combinations, 12 showed strong EDA responses (increases of at least 60% and 1μS), 6 showed weak EDA responses (increases of at least 60% but less than 1μS), and 2 showed no EDA response.

The team plans to test the hypothesis that EDA responses to focal seizures are contralateral by collecting data from both wrists [Poh, Ming-Zher, “Continuous Assessment of Epileptic Seizures with Wrist-worn Biosensors,” PhD diss., Massachusetts Institute of Technology, 2011]. If the hypothesis is proved correct stronger EDA responses will be seen on the wrist opposite the seizure focus.

The researchers also hope to present a pattern recognition algorithm later this year which further improves seizure detection accuracy for these patients and adopt device models which can be worn on the wrist. Data obtained from such devices will also help understand the causes and early signs of sudden unexpected death in patients with epilepsy.

Reference:

Cogan DL, Nourani M, Harvey J, Nagaraddi V. Abstract 3.084. Seizure detection by multi extracerebral biosignal analysis. Presented at: American Epilepsy Society Annual Meeting; Dec. 4-8, 2015; Philadelphia.

Cover image: illustrates multi extracerebral biosignal responses to a Complex Partial Seizure (left) and a Generalized Tonic Clonic Seizure (right).

Disclaimer: The International Child Neurology Association (ICNA) or the American Epilepsy Society (AES) do not endorse or recommend any particular device or technology.