Data driven determination of the level of consciousness from EEG

Projects > Machine learning based multimodal classification for Disorder of Consciousness > Projects > Data driven determination of the level of consciousness from EEG

People

• Jorge Victorino
• Francisco Gómez

Summary

Development of objective assessment tools to determine the level of consciousness in severe disabled patients (such as vegetative state/unresponsive wakefulness and minimally conscious state) is fundamental to reduce the uncertainly of behavioral evaluations []. Preliminary studies suggest that resting state EEG from these patients tend to show high levels of complexity [5,6]. This is in accordance with the information integration theory that relates consciousness to the possibility for the brain to access a wide range of states with large amounts of information [2]. These studies are focused mainly in two aspects: 1) Brain connectivity and 2) Signal complexity. In both cases, the EEG signal is represented on Fourier basis and some subbands are further studied. Importantly, this bases selection seems arbitrary and not necessarily is informative for the problem. This limitation may be reflected in the lack of sensitivity of the proposed approaches to capture the level of consciousness at individual level. In this work, we hypothesized that data driven representations adapted of the EEG resting state signal will better capture the concept of “informative” dynamics expected for high levels of consciousness. Specifically, we propose to develop a new dictionary learning sparse based representation for the EEG signal. We aim to use the “informative” concept to guide the construction of our dictionary. We will to test our hypothesis in the determination of the level of consciousness for VS/UWS and MCS patients and pharmacologically induced loss of consciousness.

Results (Expected)

• Method conference. Model proposal.
• Journal article. Complete model and application to clinical data.

References

Histopathology. 1989 Jul;15(1):49-59. Diffuse axonal injury in head injury: definition, diagnosis and grading. Adams JH, Doyle D, Ford I, Gennarelli TA, Graham DI, McLellan DR.