No abstract

No abstract

Background Evidence regarding the predictive value of early amplitude-integrated electroencephalography (aEEG)/EEG on neurodevelopmental outcomes at school age and beyond is lacking. We  aimed to investigate whether there is an association between early postnatal EEG and neurocognitive outcomes in late childhood. Methods This study is an observational prospective cohort study of premature infants with a gestational age <28 weeks. The total absolute band powers (tABP) of the delta, theta, alpha, and beta bands were analyzed from EEG recordings during the first three days of life. At 10–12 years of age, neurocognitive outcomes were assessed using the Wechsler Intelligence Scale for Children 4th edition (WISC-IV), Vineland adaptive behavior scales 2nd edition, and Behavior Rating Inventory of Executive Function (BRIEF). The mean differences in tABP were assessed for individuals with normal versus unfavorable neurocognitive scores. Results Twenty-two infants were included. tABP values in all four frequency bands were significantly lower in infants with unfavorable results in the main composite scores (full intelligence quotient, adaptive behavior composite score, and global executive composite score) on all three tests (p < 0.05). Conclusions Early postnatal EEG has the potential to assist in predicting cognitive outcomes at 10–12 years of age in extremely premature infants <28 weeks’ gestation. Impact Evidence regarding the value of early postnatal EEG in long-term prognostication in preterm infants is limited. Our study suggests that early EEG spectral analysis correlates with neurocognitive outcomes in late childhood in extremely preterm infants. Early identification of infants at-risk of later impairment is important to initiate early and targeted follow-up and intervention.

Objective: To assess and overcome the effects of site differences in EEG -based brain age prediction in preterm infants. &#xD;Approach: We used a ‘bag of features’ with a combination function estimated using support vector regression (SVR) and feature selection (filter then wrapper) to predict post-menstrual age (PMA). The SVR was trained on a dataset containing 138 EEG recordings from 37 preterm infants (site 1). A separate set of 36 EEG recordings from 36 preterm infants was used to validate the age predictor (site 2). The feature distributions were compared between sites, and training used only features that were not significantly different between sites. The mean absolute error between predicted age and PMA was used to define the accuracy of prediction. Successful validation was defined as no significant differences in error between site 1 (cross-validation) and site 2.&#xD;Main results: The age predictor based on all features and trained on site 1 was not validated on site 2 (p < 0.001; MAE site 1 = 1.0 weeks, n = 59 vs MAE site 2 = 2.1 weeks, n = 36). The MAE was improved by training on a restricted features set (MAE site 1 = 1.0 weeks, n = 59 vs MAE site 2 = 1.1 weeks, n = 36), resulting in a validated age predictor (p = 0.68). The selected features closely aligned with features selected when trained on a combination of data from site 1 and site 2.&#xD;Significance: The ability of EEG classifiers, such as brain age prediction, to maintain accuracy on data collected at other sites may be challenged by unexpected, site-dependent differences in EEG signals. Permitting a small amount of data leakage between sites improves generalization, leading towards universal methods of EEG interpretation in preterm infants.&#xD;