Prediction of Cerebral Palsy in Newborns With Hypoxic-Ischemic Encephalopathy Using Multivariate EEG Analysis and Machine Learning

Abstract

This study was carried out to investigate whether the quantitative analysis of electroencephalogram (EEG) signals of infants with hypoxic-ischemic encephalopathy (HIE) can be used for early prediction of cerebral palsy (CP). We computed sample entropy (SampEn), permutation entropy (PEn), and spectral entropy (SpEn) measures to reflect the signal's complexity and the graph-theoretic parameters derived from weighted phase-lag index (WPLI) to measure functional brain connectivity. Both feature sets were calculated in the noise-assisted multivariate empirical mode decomposition (N-A MEMD) domain to characterize the tempo-spectral integration of information and thus provide novel insight into the physiological mechanisms of the brain. Statistical analysis results showed a general deficit in the CP signals at the alpha-band component characterized by a decrease in the complexity measures and an increase in the graph-theoretic parameters specified by the diameter feature. The proposed set of features have also been evaluated using the random under-sampling boosting (RUSBoost) classifier, which was trained and tested on the feature vectors of a cohort of 26 infants -6 who developed CP by the age of 24 months and 20 with normal neuromotor outcome. A good performance of 84.6% classification accuracy (ACC), 83% sensitivity (SNS), 85% specificity (SPC) and 0.87 area under curve (AUC) was obtained using the entropy features extracted from the alpha-band component. A close result of 80.8% ACC, 67% SNS, 85% SPC and 0.79 AUC was also achieved using the diameter feature calculated from the same frequency range. Therefore, it was concluded that the obtained brain functions' characteristics successfully discriminate between the two groups of infants. These characteristics could be considered potential biomarkers of cerebral cellular damage and, therefore, could be employed in practical clinical applications for early CP prediction.INDEX TERMS Brain connectivity, cerebral palsy (CP), electroencephalogram (EEG), empirical mode decomposition (EMD), hypoxic-ischemic encephalopathy (HIE), graph theory, noise-assisted multivariate empirical mode decomposition (N-A MEMD), weighted phase-lag index (WPLI).

Principles for Optimal Window Size Selection for Infant and Adult EEG Connectivity Analysis

“…To minimize the effect of volume conduction on our connectivity estimates, we focused on two metrics, the weighted phase lag index (wPLI) [17] and weighted symbolic mutual information (wSMI) [4]. Both of them have been extensively used in EEG-based FC calculations [3], [18], [19].…”

FNIRS based study of brain network characteristics in children with cerebral palsy during bilateral lower limb movement

“…13 However, the disadvantages of MRI and DTI cannot be ignored, such as long imaging time, confined detection space, potential image distortion or blurring due to minor movements, and inability to perform real-time detection during motion tasks. Similarly, although current electroencephalography (EEG) technology has high temporal resolution, [14][15][16] the signals are easily influenced by external interference, making realtime detection of motion tasks difficult to achieve, just like MRI and DTI.…”