Edith Law combining machine and human intelligence to create crowdEEG

Wednesday, February 7, 2018

CBB member Edith Law is actively working on CrowdEEG, an innovative tool for medical time series data.

CrowdEEG is a platform that integrates machine and human intelligence for the accurate analysis of clinical EEG recordings. With the use of an open-source collaborative annotation tool, CrowdEEG enables both expert and non-expert crowds to perform feature detection or high-level classification tasks on medical time series data. Additionally, CrowdEEG implements a smart data aggregation component that uses state-of-the-art machine learning to combine insights from machines, experts and non-expert crowds for accurate analysis results on a large scale. CrowdEEG is currently focusing on specific analysis tasks, such as epileptic seizure detection, as well as sleep stage classification.

Through the development of CrowdEEG, Law seeks to prepare a high-quality dataset of human clinical polysomnograms, such as multi-channel biosignal recordings of human subjects during sleep, in order to advance other research endeavours in this field. Additionally, CrowdEEG seeks to make medical-grade EEG analysis more affordable and accessible for patients both in Canada and abroad.

More about CrowdEEG can be found here, as well as on Law's webpage.