our technology

Headache Sciences has developed a new migraine analysis method by using EEG (electroencephalography) signals to characterize migraine patients with aura (MwA).

EEG signals are used because they contain critical spatial and temporal information about the neural bioelectricity. We have developed a technique of characterizing and extracting significant, robust and informative features from EEG signals which are representative of the interictal migraine brain state. Our pilot study results show 92.9% classification accuracy of MwA in the interictal stage from the normal control (NC) group. Our findings suggest electrical predisposition of migraine which can lead to possible preventative interventions in the future. It is published in CEPHALALGIA. Vol. 37. and at Computational Science Computational Intelligence and our new preliminary 3 class (NC/MwA/MwoA) solution here CEPHALALGIA. Vol. 39.

Application of our technology


These simulations were taken while one of our patients was having a migraine compared to someone without.

The red lines show the direction and strength of delta wave currents, green is for theta, and blue is for alpha. The movie shows a spectacular complex flow of many currents. Somewhere in all the currents are the migraine signals.

Our challenge was to decipher the mystery and find the migraine signals that are buried amongst all the many other signals.

Patients journeys can involve a prolonged path it diagnosis thus delaying treatment by years. Misdiagnosis is fraught with difficulties including wrongly stigmatizing patients as having psychological problem and misdiagnosing another condition.

Although one of our many goals is to obtain FDA class II registration, at this time the test is available as an experimental consultative report.

For a detailed scientific explanation, please visit us at the poster section of 18th Congress of the International Headache Society, September 7th - 10th.

Our presentation is entitled:
"Discriminative Analysis of Migraine with Aura Using Non-Linear SVM Classification"