Quantitative Electroencephalogram in Term Neonates under Different Sleep States

doi:10.21203/rs.3.rs-2958365/v1

Download PDF

Research Article

Quantitative Electroencephalogram in Term Neonates under Different Sleep States

https://doi.org/10.21203/rs.3.rs-2958365/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 18 Oct, 2023

Read the published version in Journal of Clinical Monitoring and Computing →

You are reading this latest preprint version

Purpose:

Electroencephalogram (EEG) can be used to assess anesthetic depth, but interpreting EEG can be challenging, especially in neonates whose EEG undergo rapid changes during the perinatal course. EEG can be processed into quantitative EEG (QEEG), but limited data exist on the range of QEEG for normal term neonates during wakefulness and sleep, baseline information needed to determine changes in anesthetic depth during neonatal anesthesia. We aimed to determine the range of QEEG in neonates during awake, active sleep and quiet sleep states, and identified the ones best at discriminating between the three states.

Methods

Normal neonatal EEG from 37–46 weeks were analyzed and classified as awake, quiet sleep, or active sleep. After processing and artefact removal, total power, power ratio, coherence, and entropy, and spectral edge frequency (SEF) 50 and 90 were calculated. Descriptive statistics were used to summarize the QEEG in each of the three states. Receiver operating characteristic (ROC) curves were used to assess discriminatory ability of QEEG.

Results

30 neonates were analyzed. QEEG were different between awake vs asleep states, but similar between active vs quiet sleep states. Entropy beta, delta2 power %, coherence delta2, and SEF50 were best at discriminating awake vs active sleep. Entropy beta had the highest AUC-ROC ≥ 0.84. Entropy beta, entropy delta1, theta power %, and SEF50 were best at discriminating awake vs quiet sleep. All had AUC-ROC ≥ 0.78. In active sleep vs quiet sleep, theta power % had AUC-ROC > 0.69, lower than the other comparisons.

Conclusion

We determined the QEEG range in healthy neonates in different states of consciousness. Entropy beta and SEF50 were best at discriminating between awake and sleep states. QEEG were not as good at discriminating between quiet and active sleep. In the future, QEEG with high discriminatory power can be combinations to further improve ability to differentiate between states of consciousness.

Neonatal EEG

quantitative EEG

The electroencephalogram (EEG) is a neurodiagnostic tool that assesses brain electrical activity and has historically been used by neurologists to diagnose neurological disorders such as epilepsy. More recently, EEG monitoring has been increasingly used to assess brain development, brain injury, sleep cycles, and anesthetic depth.[1–3] However, interpreting raw EEG can be challenging for clinicians who lack formal training in EEG interpretation. This is particularly true for neonatal EEG, since rapid brain maturation contributes to continually evolving EEG characteristics (frequency, amplitude, and continuity) throughout the perinatal course.[4, 5]

Raw EEG can be computationally processed with mathematical algorithms that summarize EEG characteristics over a time epoch, yielding objective and replicable quantitative EEG (QEEG) parameters. Thus, QEEG provides an alternative to the complex, subjective, and resource-intense manual EEG interpretation.[6] In neonates, QEEG (especially amplitude integrated EEG) has been applied to assess preterm infant brain maturity,[7, 8] diagnose encephalopathy,[2] detect seizures,[9] and predict neurodevelopmental outcomes.[10] Commonly used QEEG parameters include amplitude, power spectral analysis (absolute power within a defined frequency band), relative spectral power (percent of power within a defined frequency band), entropy (signal predictability ), coherence (inter-hemispheric connectivity), and spectral edge frequency 50 or 90 (the frequency below which 50% or 90% of the activity lies under).[6]

Although neonatal QEEG has numerous advantages, limited research has been conducted on establishing the range of QEEG parameters for normal, healthy term neonates during both wakefulness and sleep. Furthermore, it is not clear which QEEG parameters are reliable indicators of the different states of consciousness. Understanding the range of QEEG parameters in healthy neonates can aid in determining changes in anesthetic depth or consciousness in this population, where frontal EEG monitoring is increasingly used during anesthesia, but reliable EEG parameters are lacking.[3, 11] We aimed to determine the range of QEEG parameters in total power, relative spectral power ratio, entropy, coherence, and spectral edge frequency in healthy full-term neonates during awake, active sleep and quiet sleep states. Additionally, we aimed to determine the QEEG parameters that most accurately discriminated between the three states.

Study Design

This retrospective study analyzed clinically obtained EEG recordings from neonates between 37 and 46 weeks adjusted gestational age at the Children’s Hospital of Philadelphia’s EEG Lab. Most neonates were referred for EEG for suspected seizure events or differential diagnosis of apneic episodes, and had EEG confirmed as normal by at least one staff pediatric electroencephalographer unrelated to this study. All EEGs were recorded on the Natus Xltek system (Natus Medical Incorp, Middleton WI. USA) with standard 10–20 lead placement and 256 Hz sampling frequency. A pediatric electroencephalographer (SLM) selected epochs of artifact-free EEG and classified the epochs as one of three states: awake, quiet sleep, or active sleep. This EEG classification was used as the gold standard for comparisons to QEEG. The average selected EEG epoch was 8.3 minutes. Some neonates did not have all three states available for analysis. The Institutional Review Board determined the study was exempt from review.

EEG Processing and Quantification

EEG processing involved: (1) selecting the bilateral frontal channels (Fp1-C3, Fp2-C4) in neonatal bipolar montage for analysis, since knowledge gained from these channels were transferable to other types of limited channel EEG monitoring (e.g., intraoperative monitoring in neonates); (2) down sampling to 64 Hz; (3) band-filtering between 0.5–30 Hz; and (4) artifact removal using a custom MATLAB (Mathworks Corp., MA, USA) program described below. The following were considered artifact: (1) continuous zero values lasting > 1 second; (2) amplitude > 1500µV; (3) same constant values > 0.1 second; or (4) impulse-like sudden changes in amplitude with a difference larger than 200µV.[4] EEG segments with artefacts identified in one channel were excluded from analysis in both channels.

After EEG processing, NEURAL,[4] an open-source and validated library of MATLAB functions developed specifically for neonatal EEG, was used to calculate QEEG parameters from Fp1-C3 and Fp2-C4. Power ratio, coherence, and entropy were quantified across five frequency bands: 0.5 ≤ delta1 < 1 Hz; 1 ≤ delta2 < 4 Hz; 4 ≤ theta < 8 Hz; 8 ≤ alpha < 13 Hz; 13 ≤ beta ≤ 30 Hz. Total power and spectral edge frequency 50 and 90 (SEF50 and SEF90) were calculated over frequency band of 0.5–30 Hz. Table 1 provides definitions of the quantification parameters. To calculate total power and power ratio, Power Spectral Density (PSD) was estimated using Welch periodogram with Hamming window of 8 second length and 50% overlap. Coherence was calculated using Pearson’s correlation between left and right paired channels. Spectral entropy was calculated using Shannon entropy and PSD as estimated above. Mathematical details on EEG quantification methods are provided in O’Toole and Boylans’ work.[4]

For each neonate, two values (from left and right frontal channels) were derived for total power, power ratio, entropy, and SEF50/90 for each of three states (awake, active sleep, and quiet sleep) over the entire EEG recording. A single coherence value was derived for each of the states over the entire EEG recording, as coherence represented the shared connectivity between the left and right channels.

Statistical Analysis

Descriptive statistics, mean and standard deviation (SD), were used to summarize the normative QEEG parameters in awake, active sleep, and quiet sleep states. Normality was assessed using histograms and Shapiro-Wilk test. To explore whether QEEG parameters were different among sleep states, repeated analysis of variance (ANOVA) measures was conducted as a global test. If the global test was significant, then post-hoc pairwise comparison was conducted with paired t-test. Bonferroni correction was used for multiple pairwise comparisons between sleep states for a total of three tests. P values from the post-hoc analyses were adjusted using Bonferroni correction.

The analysis of receiver operating characteristic (ROC) curves was conducted to assess the discriminatory ability of QEEG parameters to distinguish between states. Delong’s method was used to compute the 95% confidence interval of the area under the curve (AUC).[12] Optimal cutoff points for each QEEG parameter were determined by empirically maximizing the Youden index (sensitivity + specificity – 1).

Demographics

EEG files from 30 neonates were analyzed. The adjusted gestational age on the day of EEG recording was mean (SD): 41.8 (2.7) weeks, with 10 (33%) females. EEG was indicated for suspected seizures in 26 neonates (87%) and differential diagnosis of apneic episodes in 4 neonates (13%).

Normative QEEG values

Table 2 provides the QEEG values across the awake and sleep states. Generally, QEEG parameters were different between awake vs active and quiet sleep states, but were similar between active vs quiet sleep states.

QEEG Discriminatory Power

Figure 1 (A-F) ranks the AUC ROC of each QEEG parameter to discriminate between awake and sleep states. In awake vs active sleep (Fig. 1A-B), entropy beta, delta2 power %, coherence delta2, and SEF50 were the top discriminating QEEG parameters for both left and right channels. Entropy beta had the highest AUC ROC ≥ 0.84. In awake vs quiet sleep (Fig. 1C-D), entropy beta, entropy delta1, theta power %, and SEF50 were the top discriminating QEEG parameters for both left and right channels. Entropy beta, entropy delta1, and theta power % all had AUC ROC ≥ 0.78. In active sleep vs quiet sleep (Fig. 1E-F), theta power %, SEF90, and delta2 power % were the top discriminating QEEG parameters for both left and right channels. Theta power % had AUC ROC > 0.69, lower than the top discriminating QEEG parameters from the other two pair-wise comparisons for awake vs. active sleep and awake vs. quiet sleep.

Figure 2 (A-F) selects the QEEG parameters with AUC ROC lower 95%CI boundary > 0.5 and lists the optimal cutoff threshold for each parameter.

This study determined the range of QEEG parameters in healthy neonates in the awake, active sleep, and quiet sleep states. The QEEG parameters best at discriminating between awake and sleep states were entropy beta and SEF50, with entropy beta having the highest overall discriminatory power (AUC ROC: 0.87), similar discriminatory power (AUC ROC: 0.88) to a prior study that used different QEEG parameters.[13] In infants under anesthesia, entropy beta has been shown to decrease with increased anesthetic dose, suggesting potential commonality with decreased consciousness as seen in sleep.[14, 15]

The QEEG parameters best at discriminating between quiet sleep and active sleep were theta power % and SEF90, though both had lower discriminating power (AUC ROC: 0.66–0.69) compared to ones used to distinguish between awake and sleep states. This is different from a previous study in which active and quiet sleep could be discriminated, but not awake and sleep states.[16] That study solely relied on multi-scale permutation entropy and power spectral analysis, indicating that alternative QEEG parameters could potentially offer improved discriminatory capabilities for specific states.

QEEG analysis, which involves extracting multiple statistical and signal-processing parameters from the raw EEG, provides objective and consistent analysis.[17] This study benefited from using a combination of static (spectral power ratio) and non-linear dynamic (entropy and coherence) QEEG parameters, which yield better discrimination between awake and sleep stages.[18] Prior studies reporting normative QEEG parameters in neonates have focused on pre-term neonates,[7, 13, 19–21] EEG recorded during the first hours of life,[22] or in neonates with hypoxic-ischemic encephalopathy.[23] Few studies have reported on QEEG in healthy full term neonates,[24] especially comparing all three states.[25, 26] However, this baseline data during awake and asleep states is essential to develop QEEG parameters for intraoperative EEG monitoring in the neonatal anesthesia population. Most commercially available intraoperative EEG monitors were developed in adults and do not work well in children under one year old.[3] Furthermore, neonates are particularly sensitive to the effects of excess anesthesia so it is imperative to develop EEG algorithms that can be reliably used in this population.[27]

SEF90, the frequency below where 90% of the EEG activity/power is accumulated, decreases with decreased consciousness in older children,[11] but not in infants ≤ 2 months.[28] In one study, SEF90 was different in infants who were awake vs quiet sleep vs active sleep (9.3 vs 8.6 vs 7.1Hz, p = 0.003), though no AUC ROC values were provided to delineate discriminating power.[29] SEF90 has also been used extensively in children > 6 months as an index to assess anesthetic depth.[11, 30] SEF50, the median frequency, is used less commonly than SEF90. However, this study showed that SEF50 may discriminate between states better than SEF90 in infants less than 2 months, likely due to infants having more EEG power concentrated in lower frequencies regardless of sleep state. Therefore, during decreased consciousness associated with sleep, SEF50 may be better at detecting the smaller shift of power in the lower frequencies.

This study has several limitations. The sample size is relatively small, though similar to other studies.[16, 29] Although all neonates in this study had EEGs clinically interpreted as normal, EEGs were obtained due to a clinical concern for seizures rather than being recorded from a cohort of neonates recruited specifically to establish normative EEG data. Finally, only individual QEEG parameters were compared; combinations of high performing parameters in the future may discriminate better between states.

In conclusion, this study identified QEEG parameters that differentiated awake vs sleep states, with entropy beta having the highest overall discriminatory power. In the future, QEEG parameters with high discriminatory power can be combined to further improve ability to differentiate between states of consciousness, with the goal of developing neonatal specific EEG algorithms to assess anesthetic depth during anesthesia.

Funding:

None of the authors received funding or have disclosures related to this manuscript.

Competing Interests: 1) Ian Yuan received a research grant and speaker honorarium from Masimo Corp; 2) Matthew Kirschen received NIH funding to his institution; 3) Nicholas Abend received royalty from textbooks.

None of the above were related to this study. The other authors have no relevant financial or non-financial interests to disclose.

Results from this manuscript have been presented in abstract format at Society of Technology in Anesthesia annual conference 2022.

Author Responsibilities:

IY, CDK, JH, MPK, AAT, SSC, NSA, SLM were involved with the conception and design of the study.

IY, GG, BZ, ASH, AR, SLM were involved with data retrieval and analysis.

All authors were involved in interpretation of the data, drafting, revising the manuscript, and approved the final version of the manuscript.

H. Sawaguchi, T. Ogawa, T. Takano, and K. SATO, "Developmental changes in electroencephalogram for term and preterm infants using an autoregressive model," Pediatrics International, vol. 38, no. 6, pp. 580-589, 1996.
Y. Das, X. Wang, S. Kota, R. Zhang, H. Liu, and L. F. Chalak, "Neurovascular coupling (NVC) in newborns using processed EEG versus amplitude-EEG," Scientific reports, vol. 11, no. 1, pp. 1-7, 2021.
A. Davidson and J. Skowno, "Neuromonitoring in paediatric anaesthesia," Current Opinion in Anesthesiology, vol. 32, no. 3, pp. 370-376, 2019, doi: 10.1097/aco.0000000000000732.
J. M. Toole and G. B. Boylan, "NEURAL: quantitative features for newborn EEG using Matlab," arXiv preprint arXiv:1704.05694, 2017.
L. Cornelissen, S.-E. Kim, J. M. Lee, E. N. Brown, P. L. Purdon, and C. B. Berde, "Electroencephalographic markers of brain development during sevoflurane anaesthesia in children up to 3 years old," British journal of anaesthesia, vol. 120, no. 6, pp. 1274-1286, 2018.
J. M. O'Toole and G. B. Boylan, "Quantitative preterm EEG analysis: the need for caution in using modern data science techniques," Frontiers in Pediatrics, vol. 7, p. 174, 2019.
C. Guyer et al., "Brain maturation in the first 3 months of life, measured by electroencephalogram: A comparison between preterm and term-born infants," Clinical Neurophysiology, vol. 130, no. 10, pp. 1859-1868, 2019.
J. R. Castro Conde et al., "Assessment of neonatal EEG background and neurodevelopment in full-term small for their gestational age infants," Pediatric research, vol. 88, no. 1, pp. 91-99, 2020.
B. R. Greene, S. Faul, W. Marnane, G. Lightbody, I. Korotchikova, and G. B. Boylan, "A comparison of quantitative EEG features for neonatal seizure detection," Clinical Neurophysiology, vol. 119, no. 6, pp. 1248-1261, 2008.
E. Schumacher et al., "Automated spectral EEG analyses of premature infants during the first three days of life correlated with developmental outcomes at 24 months," Neonatology, vol. 103, no. 3, pp. 205-212, 2013.
I. Yuan, T. Xu, and C. D. Kurth, "Using electroencephalography (EEG) to guide propofol and sevoflurane dosing in pediatric anesthesia," Anesthesiology clinics, vol. 38, no. 3, pp. 709-725, 2020.
X. Robin et al., "pROC: an open-source package for R and S+ to analyze and compare ROC curves," BMC bioinformatics, vol. 12, no. 1, pp. 1-8, 2011.
N. Koolen et al., "Automated classification of neonatal sleep states using EEG," Clinical Neurophysiology, vol. 128, no. 6, pp. 1100-1108, 2017.
A. Sciusco, J. F. Standing, Y. Sheng, P. Raimondo, G. Cinnella, and M. Dambrosio, "Effect of age on the performance of bispectral and entropy indices during sevoflurane pediatric anesthesia: a pharmacometric study," Pediatric Anesthesia, vol. 27, no. 4, pp. 399-408, 2017.
Y. S. Kim, Y. J. Won, H. Jeong, B. G. Lim, M. H. Kong, and I. O. Lee, "A comparison of bispectral index and entropy during sevoflurane anesthesia induction in children with and without diplegic cerebral palsy," Entropy, vol. 21, no. 5, p. 498, 2019.
T. Wielek, R. Del Giudice, A. Lang, M. Wislowska, P. Ott, and M. Schabus, "On the development of sleep states in the first weeks of life," PloS one, vol. 14, no. 10, p. e0224521, 2019.
J. M. O’Toole, E. Pavlidis, I. Korotchikova, G. B. Boylan, and N. J. Stevenson, "Temporal evolution of quantitative EEG within 3 days of birth in early preterm infants," Scientific reports, vol. 9, no. 1, pp. 1-12, 2019.
J. Fell, J. Röschke, K. Mann, and C. Schäffner, "Discrimination of sleep stages: a comparison between spectral and nonlinear EEG measures," Electroencephalography and clinical Neurophysiology, vol. 98, no. 5, pp. 401-410, 1996.
H. J. Niemarkt et al., "Maturational changes in automated EEG spectral power analysis in preterm infants," Pediatric research, vol. 70, no. 5, pp. 529-534, 2011.
C. R. West, J. E. Harding, C. E. Williams, M. I. Gunning, and M. R. Battin, "Quantitative electroencephalographic patterns in normal preterm infants over the first week after birth," Early human development, vol. 82, no. 1, pp. 43-51, 2006.
K. Pillay, A. Dereymaeker, K. Jansen, G. Naulaers, and M. De Vos, "Applying a data-driven approach to quantify EEG maturational deviations in preterms with normal and abnormal neurodevelopmental outcomes," (in eng), Sci Rep, vol. 10, no. 1, p. 7288, Apr 29 2020, doi: 10.1038/s41598-020-64211-0.
J. R. C. Conde, D. G. Barrios, C. G. Campo, N. L. G. González, B. R. Millán, and A. J. Sosa, "Visual and Quantitative Electroencephalographic Analysis in Healthy Term Neonates Within the First Six Hours and the Third Day of Life," Pediatric Neurology, vol. 77, pp. 54-60. e1, 2017.
A. A. Garvey et al., "Multichannel EEG abnormalities during the first 6 hours in infants with mild hypoxic–ischaemic encephalopathy," Pediatric research, vol. 90, no. 1, pp. 117-124, 2021.
K. Paul, V. r. Krajča, Z. Roth, J. Melichar, and S. Petránek, "Comparison of quantitative EEG characteristics of quiet and active sleep in newborns," Sleep Medicine, vol. 4, no. 6, pp. 543-552, 2003.
A. Piryatinska, G. Terdik, W. A. Woyczynski, K. A. Loparo, M. S. Scher, and A. Zlotnik, "Automated detection of neonate EEG sleep stages," Computer methods and programs in biomedicine, vol. 95, no. 1, pp. 31-46, 2009.
M. S. Scher, J. Turnbull, K. Loparo, and M. W. Johnson, "Automated state analyses: proposed applications to neonatal neurointensive care," Journal of Clinical Neurophysiology, vol. 22, no. 4, pp. 256-270, 2005.
I. Yuan et al., "Isoelectric Electroencephalography in Infants and Toddlers during Anesthesia for Surgery: An International Observational Study," Anesthesiology, vol. 137, no. 2, pp. 187-200, 2022.
K. Hayashi, K. Shigemi, and T. Sawa, "Neonatal electroencephalography shows low sensitivity to anesthesia," Neuroscience letters, vol. 517, no. 2, pp. 87-91, 2012.
L. Bennet, K. L. Fyfe, S. R. Yiallourou, H. Merk, F. Y. Wong, and R. S. Horne, "Discrimination of sleep states using continuous cerebral bedside monitoring (amplitude‐integrated electroencephalography) compared to polysomnography in infants," Acta Paediatrica, vol. 105, no. 12, pp. e582-e587, 2016.
I. Yuan et al., "Implementation of an electroencephalogram‐guided propofol anesthesia education program in an academic pediatric anesthesia practice," Pediatric Anesthesia, vol. 32, no. 11, pp. 1252-1261, 2022.

Table 1

QEEG definitions
QEEG Parameters	Broad Definition	Detailed Definition
Power ratio over frequency bands	Power for each of the 5 frequency bands divided by total power.	Discrete Fast Fourier Transformation is applied to a segment of EEG to generate power spectrum (power vs frequency). For a given frequency band, the area under the curve of power (y axis) is divided over the entire area under curve to determine power ratio.
Coherence over frequency bands	“Synchrony” between left vs right (Fp1-C3 and Fp2-C4) channels for each of the 5 frequency bands. 0: no synchrony 1: complete synchrony	The median of Pearson's linear correlation coefficient for each pair of data points from the left and right channels.
Entropy over frequency bands	Amount of randomness in EEG signals for each of the 5 frequency bands. 0: completely predictable 1: completely random	Spectral Shannon Permutation entropy. N is the number of discrete frequencies. f (λ) = power spectral density.[25]
Spectral Edge Frequency 50/90	Frequency below which 50 or 90% of the total EEG power is accumulated.	This examples shows SEF90ED = Spectral Edge Frequency; f (λ) = power spectral density.[25]

Table 2

Quantitative EEG values for each state
Channels	QEEG Parameter		Awake (n = 29)	Active Sleep (n = 30)	Quiet Sleep (n = 21)	P value
Channels	QEEG Parameter		Awake (n = 29)	Active Sleep (n = 30)	Quiet Sleep (n = 21)	Global	Awake vs. Active Sleep	Awake vs. Quiet Sleep	Active Sleep vs Quiet Sleep
FP1-C3 (Left)	Total Power (µV²)		537.5 (452.59)	482.01 (599.4)	251.17 (199.59)	0.271
	Power Ratio (%)	Delta1	0.48 (0.18)	0.37 (0.13)	0.39 (0.1)	0.003 *	0.03 *	0.024 *	1
		Delta2	0.41 (0.12)	0.51 (0.12)	0.47 (0.08)	0.002 *	0.022 *	0.022 *	1
		Theta	0.06 (0.04)	0.08 (0.03)	0.1 (0.04)	< 0.001 *	0.077	0.01 *	0.093
		Alpha	0.02 (0.02)	0.02 (0.01)	0.02 (0.01)	0.155
		Beta	0.03 (0.03)	0.01 (0.01)	0.02 (0.01)	0.201
	Entropy	Delta1	0.94 (0.05)	0.97 (0.03)	0.98 (0.03)	0.001 *	0.045 *	0.002 *	0.105
		Delta2	0.86 (0.09)	0.9 (0.05)	0.9 (0.04)	0.003 *	0.078	0.016 *	0.579
		Theta	0.95 (0.02)	0.95 (0.02)	0.96 (0.01)	0.352
		Alpha	0.98 (0.02)	0.98 (0.01)	0.98 (0.01)	0.029 *	1	0.097	0.046 *
		Beta	0.98 (0.02)	0.94 (0.04)	0.94 (0.03)	< 0.001 *	0.001 *	< 0.001 *	0.984
	SEF50 (hz)		1.01 (0.44)	1.19 (0.3)	1.2 (0.32)	0.005 *	0.019 *	0.029 *	1
	SEF90 (hz)		4.33 (3.21)	4.02 (1.01)	4.63 (1.2)	0.459
FP2-C4 (Right)	Total Power (µV²)		399.65 (344.3)	439.98 (405.27)	226.64 (122.35)	0.07
	Power Ratio (%)	Delta1	0.47 (0.14)	0.38 (0.12)	0.37 (0.09)	0.003 *	0.058	0.014 *	1
		Delta2	0.42 (0.1)	0.51 (0.11)	0.48 (0.06)	0.004 *	0.055	0.01 *	1
		Theta	0.06 (0.03)	0.08 (0.03)	0.1 (0.04)	0.002 *	0.098	0.009 *	0.043 *
		Alpha	0.02 (0.01)	0.02 (0.01)	0.02 (0.01)	0.281
		Beta	0.03 (0.03)	0.02 (0.01)	0.03 (0.04)	0.13
	Entropy	Delta1	0.95 (0.04)	0.97 (0.04)	0.98 (0.03)	0.035 *	0.624	0.056	0.459
		Delta2	0.86 (0.07)	0.9 (0.05)	0.9 (0.04)	0.017 *	0.131	0.034 *	0.279
		Theta	0.95 (0.02)	0.96 (0.02)	0.96 (0.01)	0.225
		Alpha	0.98 (0.02)	0.98 (0.01)	0.98 (0.01)	0.732
		Beta	0.98 (0.02)	0.94 (0.04)	0.95 (0.02)	< 0.001 *	0.02 *	< 0.001 *	0.069
	SEF50 (hz)		0.98 (0.28)	1.17 (0.26)	1.21 (0.27)	0.003 *	0.03 *	0.02 *	0.75
	SEF90 (hz)		4.63 (3.59)	4.06 (1.14)	5.39 (3.59)	0.479
FP1-C3 (Left) vs FP2-C4 (Right)	Coherence	Delta1	0.39 (0.26)	0.47 (0.21)	0.41 (0.18)	0.282
		Delta2	0.37 (0.21)	0.55 (0.20)	0.48 (0.16)	0.035 *	0.118	0.253	0.792
		Theta	0.26 (0.18)	0.37 (0.25)	0.33 (0.15)	0.259
		Alpha	0.28 (0.25)	0.33 (0.31)	0.28 (0.15)	0.53
		Beta	0.3 (0.23)	0.37 (0.33)	0.21 (0.15)	0.051

Table 2: Values are shown as mean (standard deviation). Global test was conducted using repeated measures ANOVA. Pair-wise paired t-test was conducted when global test was significant. Bonferroni method was used for multiple comparisons with 3 total tests.

No competing interests reported.

Download PDF

Journal Publication

published 18 Oct, 2023

Read the published version in Journal of Clinical Monitoring and Computing →

Editorial decision: Major revision
17 Jul, 2023
Reviews received at journal
01 Jul, 2023
Reviewers agreed at journal
05 Jun, 2023
Reviewers invited by journal
25 May, 2023
Submission checks completed at journal
20 May, 2023
Editor assigned by journal
20 May, 2023
First submitted to journal
19 May, 2023

You are reading this latest preprint version

Quantitative Electroencephalogram in Term Neonates under Different Sleep States

Status:

Journal Publication

Version 1

Abstract

Purpose:

Methods

Results

Conclusion

Figures

INTRODUCTION

METHODS

Study Design

EEG Processing and Quantification

Statistical Analysis

RESULTS

Demographics

Normative QEEG values

QEEG Discriminatory Power

DISCUSSION

Declarations

References

Tables

Additional Declarations

Status:

Journal Publication

Version 1