Detection of Transient Bursts in the EEG of Preterm Infants using Time-Frequency Distributions and Machine Learning.

Short-duration bursts of spontaneous activity are important markers of maturation in the electroencephalogram (EEG) of premature infants. This paper examines the application of a feature-less machine learning approach for detecting these bursts. EEGs were recorded over the first 3 days of life for infants with a gestational age below 30 weeks. Bursts were annotated on the EEG from 36 infants. In place of feature extraction, the time-series EEG is transformed into a time-frequency distribution (TFD). A gradient boosting machine is then trained directly on the whole TFD using a leave-one-out procedure. TFD kernel parameters, length of the Doppler and lag windows, are selected within a nested cross-validation procedure during training. Results indicate that detection performance is sensitive to Doppler-window length but not lag-window length. Median area under the receiver operator characteristic for detection is 0.881 (inter-quartile range 0.850 to 0.913). Examination of feature importance highlights a critical wideband region <15 Hz in the TFD. Burst detection methods form an important component in any fully-automated brain-health index for the vulnerable preterm infant.

Detecting bursts in the EEG of very and extremely premature infants using a multi-feature approach.

AIM: To develop a method that segments preterm EEG into bursts and inter-bursts by extracting and combining multiple EEG features. METHODS: Two EEG experts annotated bursts in individual EEG channels for 36 preterm infants with gestational age < 30 weeks. The feature set included spectral, amplitude, and frequency-weighted energy features. Using a consensus annotation, feature selection removed redundant features and a support vector machine combined features. Area under the receiver operator characteristic (AUC) and Cohen's kappa (κ) evaluated performance within a cross-validation procedure. RESULTS: The proposed channel-independent method improves AUC by 4-5% over existing methods (p < 0.001, n=36), with median (95% confidence interval) AUC of 0.989 (0.973-0.997) and sensitivity-specificity of 95.8-94.4%. Agreement rates between the detector and experts' annotations, κ=0.72 (0.36-0.83) and κ=0.65 (0.32-0.81), are comparable to inter-rater agreement, κ=0.60 (0.21-0.74). CONCLUSIONS: Automating the visual identification of bursts in preterm EEG is achievable with a high level of accuracy. Multiple features, combined using a data-driven approach, improves on existing single-feature methods.

Heart rate variability in hypoxic ischemic encephalopathy during therapeutic hypothermia.

BACKGROUND: Therapeutic hypothermia (TH) aims to ameliorate further injury in infants with moderate and severe hypoxic ischemic encephalopathy (HIE). We aim to assess the effect of TH on heart rate variability (HRV) in infants with HIE. METHODS: Multichannel video-electroencephalography (EEG) and electrocardiography were assessed at 6-72 h after birth in full-term infants with HIE, recruited prior to (pre-TH group) and following (TH group) the introduction of TH in our neonatal unit. HIE severity was graded using EEG. HRV features investigated include: mean NN interval (mean NN), standard deviation of NN interval (SDNN), triangular interpolation (TINN), high-frequency (HF), low-frequency (LF), very low-frequency (VLF), and LF/HF ratio. Linear mixed model comparisons were used. RESULTS: 118 infants (pre-TH: n = 44, TH: n = 74) were assessed. The majority of HRV features decreased with increasing EEG grade. Infants with moderate HIE undergoing TH had significantly different HRV features compared with the pre-TH group (HF: P = 0.016, LF/HF ratio: P = 0.006). In the pre-TH group, LF/HF ratio was significantly different between moderate and severe HIE grades (P = 0.002). In the TH group, significant differences were observed between moderate and severe HIE grades for SDNN: P = 0.020, TINN: P = 0.005, VLF: P = 0.029, LF: P = 0.010, and HF: P = 0.006. CONCLUSION: The HF component of HRV is increased in infants with moderate HIE undergoing TH.

Heart rate variability in hypoxic ischemic encephalopathy: correlation with EEG grade and 2-y neurodevelopmental outcome.

BACKGROUND: The study aims to describe heart rate variability (HRV) in neonatal hypoxic ischemic encephalopathy (HIE) and correlate HRV with electroencephalographic (EEG) grade of HIE and neurodevelopmental outcome. METHODS: Multichannel EEG and electrocardiography (ECG) were assessed at 12-48 h after birth in healthy and encephalopathic full-term neonates. EEGs were graded (normal, mild, moderate, and severe). Neurodevelopmental outcome was assessed at 2 y of age. Seven HRV features were calculated using normalized-RR (NN) interval. The correlation of these features with EEG grade and outcome were measured using Spearman's correlation coefficient. RESULTS: HRV was significantly associated with HIE severity (P < 0.05): standard deviation of NN interval (SDNN) (r = -0.62), triangular interpolation of NN interval histogram (TINN) (r = -0.65), mean NN interval (r = -0.48), and the very low frequency (VLF) (r = -0.60), low frequency (LF) (r = -0.67) and high frequency (HF) components of the NN interval (r = -0.60). SDNN at 24 and 48 h were significantly associated (P < 0.05) with neurodevelopmental outcome (r = -0.41 and -0.54, respectively). CONCLUSION: HRV is associated with EEG grade of HIE and neurodevelopmental outcome. HRV has potential as a prognostic tool to complement EEG.

Automated analysis of multi-channel EEG in preterm infants.

OBJECTIVE: To develop and validate two automatic methods for the detection of burst and interburst periods in preterm eight-channel electroencephalographs (EEG). To perform a detailed analysis of interobserver agreement on burst and interburst periods and use this as a benchmark for the performance of the automatic methods. To examine mathematical features of the EEG signal and their potential correlation with gestational age. METHODS: Multi-channel EEG from 36 infants, born at less than 30 weeks gestation was utilised, with a 10 min artifact-free epoch selected for each subject. Three independent expert observers annotated all EEG activity bursts in the dataset. Two automatic algorithms for burst/interburst detection were applied to the EEG data and their performances were analysed and compared with interobserver agreement. A total of 12 mathematical features of the EEG signal were calculated and correlated with gestational age. RESULTS: The mean interobserver agreement was found to be 77% while mean algorithm/observer agreement was 81%. Six of the mathematical features calculated (spectral entropy, Higuchi fractal dimension, spectral edge frequency, variance, extrema median and Hilberts transform amplitude) were found to have significant correlation with gestational age. CONCLUSIONS: Automatic detection of burst/interburst periods has been performed in multi-channel EEG of 36 preterm infants. The algorithm agreement with expert observers is found to be on a par with interobserver agreement. Mathematical features of EEG have been calculated which show significant correlation with gestational age. SIGNIFICANCE: Automatic analysis of preterm multi-channel EEG is possible. The methods described here have the potential to be incorporated into a fully automatic system to quantitatively assess brain maturity from preterm EEG.