Seizure detection accuracy


We are constantly validating and improving our algorithm, to test its accuracy we collaborated with top hospitals to collect clinically labelled seizure data using EEG (v-EEG). The data consisted of 192 recordings taken from 53 patients wearing an Empatica E3 or E4 wrist sensor recording EDA and accelerometer data.

The data were analyzed using our software (Empatica, Inc.). The algorithm tested achieved a seizure detection rate for GTC (general tonic-clonic) of 100% on 12 seizures, while maintaining a false alarm rate of approximately one per day over 164 days. We presented these findings at the Partners Against Mortality in Epilepsy (PAME) Meeting in Virginia on June 23-26, 2016.

For the Annual Epilepsy Meeting in Prague on September 11-15, 2016, we presented further findings using a different data set.  Using a larger population sample, there was on average 1 false alarm per day with  90% sensitivity.  We had on average 2 false alarms per day when we increased the sensitivity to 95%, and 5.7 when we extended to 100% sensitivity. This shows that the more sensitive our algorithm is, the higher likelihood it has of triggering false alarms.


Have more questions?