TY - JOUR
T1 - Multi-site EEG studies in early infancy
T2 - Methods to enhance data quality
AU - for the IBIS Network
AU - Dickinson, Abigail
AU - Booth, Madison
AU - Daniel, Manjari
AU - Campbell, Alana
AU - Miller, Neely
AU - Lau, Bonnie
AU - Zempel, John
AU - Webb, Sara Jane
AU - Elison, Jed
AU - Lee, Adrian K.C.
AU - Estes, Annette
AU - Dager, Stephen
AU - Hazlett, Heather
AU - Wolff, Jason
AU - Schultz, Robert
AU - Marrus, Natasha
AU - Evans, Alan
AU - Piven, Joseph
AU - Pruett, John R.
AU - Jeste, Shafali
N1 - Publisher Copyright:
© 2024 The Authors
PY - 2024/10
Y1 - 2024/10
N2 - Brain differences linked to autism spectrum disorder (ASD) can manifest before observable symptoms. Studying these early neural precursors in larger and more diverse cohorts is crucial for advancing our understanding of developmental pathways and potentially facilitating earlier identification. EEG is an ideal tool for investigating early neural differences in ASD, given its scalability and high tolerability in infant populations. In this context, we integrated EEG into an existing multi-site MRI study of infants with a higher familial likelihood of developing ASD. This paper describes the comprehensive protocol established to collect longitudinal, high-density EEG data from infants across five sites as part of the Infant Brain Imaging Study (IBIS) Network and reports interim feasibility and data quality results. We evaluated feasibility by measuring the percentage of infants from whom we successfully collected each EEG paradigm. The quality of task-free data was assessed based on the duration of EEG recordings remaining after artifact removal. Preliminary analyses revealed low data loss, with average in-session loss rates at 4.16 % and quality control loss rates at 11.66 %. Overall, the task-free data retention rate, accounting for both in-session issues and quality control, was 84.16 %, with high consistency across sites. The insights gained from this preliminary analysis highlight key sources of data attrition and provide practical considerations to guide similar research endeavors.
AB - Brain differences linked to autism spectrum disorder (ASD) can manifest before observable symptoms. Studying these early neural precursors in larger and more diverse cohorts is crucial for advancing our understanding of developmental pathways and potentially facilitating earlier identification. EEG is an ideal tool for investigating early neural differences in ASD, given its scalability and high tolerability in infant populations. In this context, we integrated EEG into an existing multi-site MRI study of infants with a higher familial likelihood of developing ASD. This paper describes the comprehensive protocol established to collect longitudinal, high-density EEG data from infants across five sites as part of the Infant Brain Imaging Study (IBIS) Network and reports interim feasibility and data quality results. We evaluated feasibility by measuring the percentage of infants from whom we successfully collected each EEG paradigm. The quality of task-free data was assessed based on the duration of EEG recordings remaining after artifact removal. Preliminary analyses revealed low data loss, with average in-session loss rates at 4.16 % and quality control loss rates at 11.66 %. Overall, the task-free data retention rate, accounting for both in-session issues and quality control, was 84.16 %, with high consistency across sites. The insights gained from this preliminary analysis highlight key sources of data attrition and provide practical considerations to guide similar research endeavors.
KW - Autism
KW - Early identification
KW - Electrophysiology
KW - Multi-site
KW - Multimodal
UR - http://www.scopus.com/inward/record.url?scp=85201494959&partnerID=8YFLogxK
U2 - 10.1016/j.dcn.2024.101425
DO - 10.1016/j.dcn.2024.101425
M3 - Article
C2 - 39163782
AN - SCOPUS:85201494959
SN - 1878-9293
VL - 69
JO - Developmental Cognitive Neuroscience
JF - Developmental Cognitive Neuroscience
M1 - 101425
ER -