Predictive Modeling from High-Dimensional MEG

Python pipelines from 306-channel MEG time series to predictive models of language and reading outcomes across pediatric cohorts.

Role: Research Science Engineer, I-LABS, University of Washington (2011-2023)

Highlights:

• Infant vocabulary prediction (r = 0.73, p = 0.001)
• Reading-skill biomarker (r = 0.67, p = 1x10^-6)
• Linear mixed effects, cluster permutation testing, dimensionality reduction

Outcomes: Developmental Cognitive Neuroscience (2021); Brain and Language (2021)

Python MNE-Python MATLAB advanced statistics MEG