Poster No:
1024
Submission Type:
Abstract Submission
Authors:
Nathan Stevenson1, Sampsa Vanhatalo2, Maria Velilla3, Ana Alarcón3, James Roberts4
Institutions:
1QIMR Berghofer, Brisbane, QLD, 2University of Helsinki, Helsinki, Uusimaa, 3Hospital Sant Joan de Déu Barcelona, Barcelona, Spain, 4QIMR Berghofer Medical Research Institute, Brisbane, QLD
First Author:
Co-Author(s):
Maria Velilla
Hospital Sant Joan de Déu Barcelona
Barcelona, Spain
Ana Alarcón
Hospital Sant Joan de Déu Barcelona
Barcelona, Spain
James Roberts
QIMR Berghofer Medical Research Institute
Brisbane, QLD
Introduction:
Surrogate measures of biological age interpreted with respect to actual age have shown promise as a biomarkers of human health. In preterm infants, where neurodevelopment is a fundamental concern, measures of brain age are under development. Here, we examine the role of confounders on predictions of age derived from limited channel recordings of the electroencephalogram.
Methods:
A dataset of 449, two-channel, serial EEG recordings from 94 preterm infants was used (after the rejection of 35 recordings due to poor quality). The typical duration of EEG recordings was 3h. A set of 49 features were extracted from 1h epochs of EEG recordings (overlap of 30 minutes) and summarized across two-channels (mean) and extracted epochs (median). The post-menstrual age (PMA) range of recordings was 24-38 weeks. Functional brain age was predicted with a subset of features (feature selection) fed into a support vector regression with nonlinear kernel (Gaussian). Data augmentation ensured balance between data samples and PMA. Age was evaluated within a 10-fold cross-validation using mean absolute error (MAE) and the correlation coefficient between predicted age and post-menstrual age. The effect of confounders such as age, age squared, neurodevelopmental outcome, sedation, small for gestational age (SGA) and sex on age prediction were examined using linear regression.
Results:
Age prediction accuracy was high with a MAE of 0.84 weeks (95%CI: 0.77 to 0.92; n = 449) and an r² of 0.840 (95%CI: 0.811 to 0.865; n = 449). Analysis of a linear regression of confounders with predicted age showed that age, age squared, being SGA, sedated or having EEG affecting pathology were significant contributors to prediction variance (p < 0.05). Neurodevelopmental outcome and biological sex were not.
Conclusions:
Accurate interpretation of brain age is dependent on factors such as birthweight, sedation and pathology. Consideration of these variables is, therefore, necessary when using predicted age difference as a biomarker of brain health.
Disorders of the Nervous System:
Neurodevelopmental/ Early Life (eg. ADHD, autism)
Lifespan Development:
Normal Brain Development: Fetus to Adolescence 1
Novel Imaging Acquisition Methods:
EEG 2
Keywords:
Electroencephaolography (EEG)
Machine Learning
1|2Indicates the priority used for review
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
I accept
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information.
Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
I do not want to participate in the reproducibility challenge.
Please indicate below if your study was a "resting state" or "task-activation” study.
Other
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Patients
Was this research conducted in the United States?
No
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel?
NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Yes
Were any animal research approved by the relevant IACUC or other animal research panel?
NOTE: Any animal studies without IACUC approval will be automatically rejected.
Not applicable
Please indicate which methods were used in your research:
EEG/ERP
Provide references using APA citation style.
N/A
No