A machine learning artefact detection method for single-channel infant event-related potential studies.

Objective. Automated detection of artefact in stimulus-evoked electroencephalographic (EEG) data recorded in neonates will improve the reproducibility and speed of analysis in clinical research compared with manual identification of artefact. Some studies use very short, single-channel epochs of EEG data with little recorded EEG per infant-for example because the clinical vulnerability of the infants limits access for recording. Current artefact-detection methods that perform well on adult data and resting-state and multi-channel data in infants are not suitable for this application. The aim of this study was to create and test an automated method of detecting artefact in single-channel 1500 ms epochs of infant EEG.Approach. A total of 410 epochs of EEG were used, collected from 160 infants of 28-43 weeks postmenstrual age. This dataset-which was balanced to include epochs of background activity and responses to visual, auditory, tactile and noxious stimuli-was presented to seven independent raters, who independently labelled the epochs according to whether or not they were able to visually identify artefacts. The data was split into a training set (340 epochs) and an independent test set (70 epochs). A random forest model was trained to identify epochs as either artefact or not artefact.Main results. This model performs well, achieving a balanced accuracy of 0.81, which is as good as manual review of data. Accuracy was not significantly related to the infant age or type of stimulus.Significance. This method provides an objective tool for automated artefact rejection for short epoch, single-channel EEG in neonates and could increase the utility of EEG in neonates in both the clinical and research setting.

Effect of parental touch on relieving acute procedural pain in neonates and parental anxiety (Petal): a multicentre, randomised controlled trial in the UK.

BACKGROUND: Touch interventions such as massage and skin-to-skin contact relieve neonatal pain. The Parental touch trial (Petal) aimed to assess whether parental stroking of their baby before a clinically required heel lance, at a speed of approximately 3 cm/s to optimally activate C-tactile nerve fibres, provides effective pain relief. METHODS: Petal is a multicentre, randomised, parallel-group interventional superiority trial conducted in the John Radcliffe Hospital (Oxford University Hospitals NHS Foundation Trust, Oxford, UK) and the Royal Devon and Exeter Hospital (Royal Devon University Healthcare NHS Foundation Trust, Exeter, UK). Neonates without neurological abnormalities who were born at 35 weeks gestational age or more and required a blood test via a heel lance in the first week of life were randomly assigned (1:1) to receive parental touch for 10 s either before (intervention group) or after (control group) the clinically required heel lance. Randomisation was managed at the Oxford site using a web-based minimisation algorithm with allocation concealment. The primary outcome measure was the magnitude of noxious-evoked brain activity in response to the heel lance measured with electroencephalography (EEG). Secondary outcome measures were Premature Infant Pain Profile-Revised (PIPP-R) score, development of tachycardia, and parental anxiety score. For all outcomes, the per-protocol effect was estimated via complier average causal effect analysis on the full analysis set. The trial is registered on ISRCTN (ISRCTN14135962) and ClinicalTrials.gov (NCT04901611). FINDINGS: Between Sept 1, 2021, and Feb 7, 2023, 159 parents were approached to participate in the study, and 112 neonates were included. 56 neonates were randomly assigned to the intervention group of parental stroking before the heel lance and 56 to the control group of parental stroking after the heel lance. The mean of the magnitude of the heel lance-evoked brain activity was 0·85 arbitrary units (a.u.; SD 0·70; n=39; a scaled magnitude of 1 a.u. represents the expected mean response to a heel lance in term-aged neonates) in the intervention group and 0·91 a.u. (SD 0·76; n=43) in the control group. Therefore, the primary outcome did not differ significantly between groups, with a mean difference of -0·11 a.u. (lower in intervention group; SD 0·77; 95% CI -0·42 to 0·20; p=0·38; n=82). No significant difference was observed across secondary outcomes. The PIPP-R difference in means was 1·10 (higher in intervention group, 95% CI -0·42 to 2·61; p=0·15; n=100); the odds ratio of becoming tachycardic was 2·08 (95% CI 0·46 to 9·46; p=0·34, n=105) in the intervention group with reference to the control group; and the difference in parental State-Trait Anxiety Inventory-State score was -0·44 (higher in control group; SD 6·85; 95% CI -2·91 to 2·02; p=0·72; n=106). One serious adverse event (desaturation) occurred in a neonate randomly assigned to the control group, which was not considered to be related to the study. INTERPRETATION: Parental stroking delivered at an optimal speed to activate C-tactile fibres for a duration of 10 s before the painful procedure did not significantly change neonates' magnitude of pain-related brain activity, PIPP-R score, or development of tachycardia. The trial highlighted the challenge of translating an experimental researcher-led tactile intervention into a parent-led approach, and the value of involving parents in their baby's pain management. FUNDING: Wellcome Trust and Bliss.

The effect of acute respiratory events and respiratory stimulants on EEG-recorded brain activity in neonates: A systematic review.

Objective: We conducted a systematic review to investigate electroencephalography (EEG) changes during periods of acute respiratory events such as apnoea and the effect of respiratory stimulants on EEG features in infants. Methods: Studies examining respiration and EEG-recorded brain activity in human neonates between 28 and 42 weeks postmenstrual age were included. Two reviewers independently screened all records and included studies were assessed using the Joanna Briggs Institute Critical Appraisal Tool. The protocol was registered in PROSPERO (CRD42022339873). Results: We identified 14 studies with a total of 534 infants. Nine articles assessed EEG changes in relation to apnoea, one assessed hiccups, and four investigated the effect of respiratory stimulants. The relationship between neonatal apnoea and EEG changes was inconsistent; EEG suppression and decreased amplitude and frequency were observed during some, but not all, apnoeas. Respiratory stimulants increased EEG continuity compared with before use. Conclusions: Current studies in this area are constrained by small sample sizes. Diverse exposure definitions and outcome measures impact inference. Significance: This review highlights the need for further work; understanding the relationship between respiration and the developing brain is key to mitigating the long-term effects of apnoea.

Effects of timing parameter changes on the gait of functional electrical stimulation users with drop foot.

INTRODUCTION: Functional electrical stimulation uses clinician-set parameters to modify stimulation. This study aimed to investigate whether timing parameters in the ODFS Pace functional electrical stimulation device have an effect on the gait of the general population of functional electrical stimulation users who have a foot drop. METHODS: Twelve functional electrical stimulation users with foot drop resulting from upper motor neurone disorders were recruited from the functional electrical stimulation Service in Leeds, UK. A crossover trial design was used, comparing adjusted values of rising ramp, delay and extension. Instrumented gait analysis was carried out to measure ankle dorsiflexion during the swing phase of gait, foot clearance from the ground, and speed of ankle plantarflexion at initial contact. The effect of timing parameters on gait kinematics was studied. RESULTS: No statistically significant effects on the measured parts of gait were found for any of the timing parameters. Trends were identified in average mid-swing ground clearance and dorsiflexion associated with the delay and rising ramp timing parameters. CONCLUSIONS: Further work in this area should use larger numbers of participants. Based on these results, the effects of ramping and delay would be of particular interest for further study.