home concepts publications press




Automatic detection of convulsive seizures using a wearable sensor.

The sudden, unforseen way in which epileptic seizures tend to strike is one of the most disabling aspects of the disorder. Since seizures often cause the loss of consciousness, the ability to detect a seizure event is important for timely treatment. How can we design a seizure detector that is suitable for everyday use in terms of cost, comfort and social acceptance?

I developed an algorithm for automatic detection of convulsive seizures based on electrodermal activity and actigraphy using a wrist-worn biosensor. The new seizure detection algorithm was tested on >4,213 h of recordings from 80 patients and detected 15 (94%) of 16 of the GTC seizures from seven patients with 130 false alarms (0.74 per 24 h). This algorithm can potentially provide a convulsive seizure alarm system for caregivers and objective quantification of seizure frequency.


M.Z. Poh, T. Loddenkemper, N.C. Swenson, S. Goyal, M. Sabtala, J.R. Madsen and R.W. Picard.
"Convulsive seizure detection using a wrist-worn electrodermal activity and accelerometry biosensor."
Epilepsia, vol. 53, no. 5, e93-e97, 2012.
[pdf] [supplementary materials]



Example of a 24-hour recording using the wrist-worn biosensor.


Seizure detection methodology.


Wrist-worn biosensor.


Patient resting in bed.


Patient playing with a ball.


home | research | publications | press | vita

Copyright © 2013 Ming-Zher Poh