Grant: Discovery Grants Program - Individual
Université de Montréal
$37,000.00
Cross-frequency coupling and machine learning for electrophysiological signal analysis
FY 2023-24
Summary
Electroencephalography (EEG) offers a direct window into the brain's dynamic activity. EEG signals are continuous multichannel time series which record the brain's complex electrical activity through electrode sensors. Over the last decade, the development of mass storage technologies and improved high-quality acquisition devices have driven an exponential increase in the amount of stored biological data including clinical EEG. To improve our understanding of biological systems, efficient computational tools to analyze such data sets should be developed and refined. Indeed, EEG data can be used as input to artificial intelligence algorithms to automate the analysis and study of the brain or to improve our understanding of the brain's physiological and pathological mechanisms. Computational signal processing techniques do not only accelerate the analysis of large quantities of data but also uncover hidden brain mechanisms underlying physiological and pathological activity. With this research program, we aim to develop an open-source technology that will help researchers quantitatively study brain activity as recorded with EEG. The long-term goal of my research program is: To establish novel quantitative measures and artificial intelligence algorithms for the analysis of EEG recordings Specifically, we will: 1) design a novel approach for automatic feature extraction and classification of EEG signals based on nonlinear activity and machine learning, 2) develop a novel brain network analysis framework based on multifrequency connectivity within and between brain regions, and 3) incorporate developed techniques into a platform for automated feature extraction, classification, and brain connectivity analysis. The proposed techniques will be validated on synthetic signals and tested on human EEG recordings for the classification of epileptic brain states and for the quantitative delineation of the seizure-generating brain region. Findings from this program will provide novel automated signal processing tools to the scientific research community for feature extraction, classification, and network analysis of electrophysiological signals. An avant-garde web-based platform will facilitate the transfer of proposed methodologies to the scientific community. Beyond improving our understanding of brain mechanisms via extensive EEG analysis, our proposed methodologies can be easily adapted for the analysis of other electrophysiological time series such as electrocardiography and electromyography. Our novel algorithms will contribute to a variety of brain machine interface applications (e.g., motor imagery classification, neuroprosthetics) from human and non-human electrophysiological signals (e.g, non-human primates, rodents).
Program
Discovery Grants Program - Individual
Recipient
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
RGPIN-2023-04189
Name nom
Bou Assi, Elie
Department département
Neurosciences
Organization id
63
Institution établissement
Université de Montréal
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2023
Competition year année de concours
2023
Award amount
37000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1510
Committee name en
Electrical and Computer Engineering
Committee name fr
Génie électrique et informatique
Area of application code
1102
Area of application group en
Health, education and social services
Area of application group fr
Santé publique, éducation et services sociaux
Area of application en
Biomedical engineering
Area of application fr
Génie biomédical
Research subject code
2514
Research subject group en
Electrical and electronic engineering
Research subject group fr
Digital signal processing
Research subject en
Génie électrique et électronique
Research subject fr
Traitement des signaux numériques
Application title
Cross-frequency coupling and machine learning for electrophysiological signal analysis
Keyword
Automatic feature extraction; Brain connectivity; Cross frequency coupling; Electroencephalography; Higher order spectra; Machine learning; Signal processing
Application summary
Electroencephalography (EEG) offers a direct window into the brain's dynamic activity. EEG signals are continuous multichannel time series which record the brain's complex electrical activity through electrode sensors. Over the last decade, the development of mass storage technologies and improved high-quality acquisition devices have driven an exponential increase in the amount of stored biological data including clinical EEG. To improve our understanding of biological systems, efficient computational tools to analyze such data sets should be developed and refined. Indeed, EEG data can be used as input to artificial intelligence algorithms to automate the analysis and study of the brain or to improve our understanding of the brain's physiological and pathological mechanisms. Computational signal processing techniques do not only accelerate the analysis of large quantities of data but also uncover hidden brain mechanisms underlying physiological and pathological activity. With this research program, we aim to develop an open-source technology that will help researchers quantitatively study brain activity as recorded with EEG. The long-term goal of my research program is: To establish novel quantitative measures and artificial intelligence algorithms for the analysis of EEG recordings Specifically, we will: 1) design a novel approach for automatic feature extraction and classification of EEG signals based on nonlinear activity and machine learning, 2) develop a novel brain network analysis framework based on multifrequency connectivity within and between brain regions, and 3) incorporate developed techniques into a platform for automated feature extraction, classification, and brain connectivity analysis. The proposed techniques will be validated on synthetic signals and tested on human EEG recordings for the classification of epileptic brain states and for the quantitative delineation of the seizure-generating brain region. Findings from this program will provide novel automated signal processing tools to the scientific research community for feature extraction, classification, and network analysis of electrophysiological signals. An avant-garde web-based platform will facilitate the transfer of proposed methodologies to the scientific community. Beyond improving our understanding of brain mechanisms via extensive EEG analysis, our proposed methodologies can be easily adapted for the analysis of other electrophysiological time series such as electrocardiography and electromyography. Our novel algorithms will contribute to a variety of brain machine interface applications (e.g., motor imagery classification, neuroprosthetics) from human and non-human electrophysiological signals (e.g, non-human primates, rodents).