Several National Research Council of Canada and academic grants to use software, simulations and machine learning for epilepsy diagnosis and fatigue studies.
Client Since
1991
Industry
Academic
Government
Services
Software Development
Simulations
Synopsis
Praxis used neural networks and machine learning to process and categorize EEG data.
The Project
Praxis had a long and established relationship with the Brain Behavior Lab at Simon Fraser University. The projects were driven by academic research into areas of mutual interest. Initial work centered around fatigue studies conducted on airline pilots and air traffic controllers using software and simulations developed by Praxis. The Brain Behavior Lab later approached Praxis to assist in a study on epilepsy in children. Epilepsy in children posed significant diagnostic challenges at the time, due to the difficult in distinguishing between benign and non-benign forms. The subtleties of Electroencephalograms (EEGs), used in diagnosis, often confounded even experienced EEG readers, making correct diagnosis a difficult but critical problem.
The Result
Praxis partnered with the Brain Behavior Lab at Simon Fraser University to employ machine learning early Artificial Intelligence (AI) for epilepsy diagnosis. By developing a multi-level neural network trained on verified epilepsy data, Praxis’s system was able to categorize unexamined EEGs into benign and non-benign groups.
Praxis later went on to apply these same machine learning neural networks to create process simulations. By training the systems on process data collected from existing process units.
