Artificial intelligence helps scientists to measure human consciousness
New research led by brain and computer scientists from the Universities of Korea and Liège and published in the prestigious journal ‘Nature Communications’ shows Artificial Intelligence (AI) offers unique insights in quantifying changes in consciousness during sleep, dreaming, anesthesia and coma.
Understanding and measuring the human mind remains one of science’s biggest challenges and has major medical, societal, and ethical impact. Currently, healthcare workers mostly make subjective interpretations of a patient’s reactivity and motor movements to appreciate consciousness during surgery under anesthesia or when treating brain-damaged patients in coma and related states. Previous research by the GIGA Consciousness Research Unit and Coma Science Group at the University and University Hospital of Liège founded by Prof Steven Laureys had already shown that clinicians may often underestimate the possibility of residual conscious awareness in these conditions.
Here, the team of Prof Seong-Whan Lee, Dept of Artificial Intelligence from Korea University, in collaboration with the team of Dr Olivia Gosseries, co-director of the Belgian Coma Science Group, used AI deep learning to develop novel and unique markers of both the level and content of consciousness by analysing electrical measures of brain function in >100 human participants studied during different conditions of modified consciousness (including propofol and ketamine sedation but also severe traumatic injury, stroke and cardiac arrest survivors).
“This paper is the world's first technology to quantify the arousal and awareness at the same time" said Dr Seong-Whan. Dr. Gosseries added that “future studies are needed to implement this novel indicator in the clinical routine and even to develop an online and real-time tool to be implemented in our hospitals, operating rooms and intensive care units”.
This study was funded by FNRS, Fund Generate of the King Baudouin Foundation, Human Brain Project, Fondazione Europea di Ricerca Biomedica, BIAL Foundation, AstraZeneca Foundation, James McDonnell Foundation, Mind Science Foundation, Tiny Blue Dot Foundation, National Institutes of Health, Academy of Finland and the Korean government.
"Quantifying arousal and awareness in altered states of consciousness using interpretable deep learning" by Minji Lee, Leandro Sanz, Alice Barra, Audrey Wolff, Jaakko Nieminen, Melanie Boly, Mario Rosanova, Silvia Casarotto, Olivier Bodart, Jitka Annen, Aurore Thibaut, Rajanikant Panda, Vincent Bonhomme, Marcello Massimini, Giulio Tononi, Steven Laureys, Olivia Gosseries*, and Seong-Whan Lee*, Nature Communications
*These authors contributed equally
Dept of Artificial Intelligence and Brain & Cognitive Engineering, Korea U; Coma Science Group, GIGA-Consciousness, U Liège; Centre du Cerveau and Anesthesia Dept, U Hospital of Liège; Wisconsin Institute for Sleep & Consciousness, U Wisconsin, Madison; Dept Neuroscience & Biomedical Engineering, Aalto U School of Science, Espoo, Finland; Dept Biomedical & Clinical Sciences U Milan, Italy.