EEG Complexity as a Biomarker for Autism Risk
Author Information
Author(s): Williamb Bosl, Atrienne Tierney, Helen Tager-Flusberg, Charles Nelson
Primary Institution: Harvard Medical School
Hypothesis
Can modified multiscale entropy (mMSE) of resting state EEG data distinguish typically developing children from infants at high risk for autism spectrum disorder?
Conclusion
The study suggests that mMSE from resting state EEG signals may be a useful biomarker for early detection of autism risk in infants.
Supporting Evidence
- Infants at high risk for autism showed different EEG complexity patterns compared to typically developing infants.
- Classification accuracy for boys was close to 100% at age 9 months.
- Significant differences in mMSE were found at most ages except 9 months.
Takeaway
This study looks at brain waves in babies to see if they can tell if a baby might have autism later on. It found that certain patterns in the brain waves can help spot babies at risk.
Methodology
The study used modified multiscale entropy (mMSE) computed from EEG data and classified infants using a multiclass support vector machine algorithm.
Potential Biases
Potential bias due to the small sample size and the reliance on EEG data from a specific age range.
Limitations
The study treated data from multiple sessions of the same infant as independent, which may not account for individual developmental trajectories.
Participant Demographics
79 infants, 46 at high risk for autism (HRA) and 33 controls, aged 6 to 24 months.
Statistical Information
P-Value
0.01
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website