Dataset Information

A machine-learning algorithm for neonatal seizure recognition: a multicentre, randomised, controlled trial.

ABSTRACT: BACKGROUND:Despite the availability of continuous conventional electroencephalography (cEEG), accurate diagnosis of neonatal seizures is challenging in clinical practice. Algorithms for decision support in the recognition of neonatal seizures could improve detection. We aimed to assess the diagnostic accuracy of an automated seizure detection algorithm called Algorithm for Neonatal Seizure Recognition (ANSeR). METHODS:This multicentre, randomised, two-arm, parallel, controlled trial was done in eight neonatal centres across Ireland, the Netherlands, Sweden, and the UK. Neonates with a corrected gestational age between 36 and 44 weeks with, or at significant risk of, seizures requiring EEG monitoring, received cEEG plus ANSeR linked to the EEG monitor displaying a seizure probability trend in real time (algorithm group) or cEEG monitoring alone (non-algorithm group). The primary outcome was diagnostic accuracy (sensitivity, specificity, and false detection rate) of health-care professionals to identify neonates with electrographic seizures and seizure hours with and without the support of the ANSeR algorithm. Neonates with data on the outcome of interest were included in the analysis. This study is registered with ClinicalTrials.gov, NCT02431780. FINDINGS:Between Feb 13, 2015, and Feb 7, 2017, 132 neonates were randomly assigned to the algorithm group and 132 to the non-algorithm group. Six neonates were excluded (four from the algorithm group and two from the non-algorithm group). Electrographic seizures were present in 32 (25·0%) of 128 neonates in the algorithm group and 38 (29·2%) of 130 neonates in the non-algorithm group. For recognition of neonates with electrographic seizures, sensitivity was 81·3% (95% CI 66·7-93·3) in the algorithm group and 89·5% (78·4-97·5) in the non-algorithm group; specificity was 84·4% (95% CI 76·9-91·0) in the algorithm group and 89·1% (82·5-94·7) in the non-algorithm group; and the false detection rate was 36·6% (95% CI 22·7-52·1) in the algorithm group and 22·7% (11·6-35·9) in the non-algorithm group. We identified 659 h in which seizures occurred (seizure hours): 268 h in the algorithm versus 391 h in the non-algorithm group. The percentage of seizure hours correctly identified was higher in the algorithm group than in the non-algorithm group (177 [66·0%; 95% CI 53·8-77·3] of 268 h vs 177 [45·3%; 34·5-58·3] of 391 h; difference 20·8% [3·6-37·1]). No significant differences were seen in the percentage of neonates with seizures given at least one inappropriate antiseizure medication (37·5% [95% CI 25·0 to 56·3] vs 31·6% [21·1 to 47·4]; difference 5·9% [-14·0 to 26·3]). INTERPRETATION:ANSeR, a machine-learning algorithm, is safe and able to accurately detect neonatal seizures. Although the algorithm did not enhance identification of individual neonates with seizures beyond conventional EEG, recognition of seizure hours was improved with use of ANSeR. The benefit might be greater in less experienced centres, but further study is required. FUNDING:Wellcome Trust, Science Foundation Ireland, and Nihon Kohden.

SUBMITTER: Pavel AM

PROVIDER: S-EPMC7492960 | biostudies-literature | 2020 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A machine-learning algorithm for neonatal seizure recognition: a multicentre, randomised, controlled trial.

Pavel Andreea M AM Rennie Janet M JM de Vries Linda S LS Blennow Mats M Foran Adrienne A Shah Divyen K DK Pressler Ronit M RM Kapellou Olga O Dempsey Eugene M EM Mathieson Sean R SR Pavlidis Elena E van Huffelen Alexander C AC Livingstone Vicki V Toet Mona C MC Weeke Lauren C LC Finder Mikael M Mitra Subhabrata S Murray Deirdre M DM Marnane William P WP Boylan Geraldine B GB

The Lancet. Child & adolescent health 20200827 10

<h4>Background</h4>Despite the availability of continuous conventional electroencephalography (cEEG), accurate diagnosis of neonatal seizures is challenging in clinical practice. Algorithms for decision support in the recognition of neonatal seizures could improve detection. We aimed to assess the diagnostic accuracy of an automated seizure detection algorithm called Algorithm for Neonatal Seizure Recognition (ANSeR).<h4>Methods</h4>This multicentre, randomised, two-arm, parallel, controlled tri ...[more]

PMID: 32861271

Similar Datasets

Project description:BackgroundSepsis is an important cause of neonatal death and perinatal brain damage, particularly in preterm infants. While effective antibiotic treatment is essential treatment for sepsis, resistance to antibiotics is increasing. Adjuvant therapies, such as intravenous immunoglobulin, therefore offer an important additional strategy. Three Cochrane systematic reviews of randomised controlled trials in nearly 6,000 patients suggest that non-specific, polyclonal intravenous immunoglobulin is safe and reduces sepsis by about 15% when used as prophylaxis but does not reduce mortality in this situation. When intravenous immunoglobulin is used in the acute treatment of neonatal sepsis, however, there is a suggestion that it may reduce mortality by 45%. However, the existing trials of treatment were small and lacked long-term follow-up data.This study will assess reliably whether treatment of neonatal sepsis with intravenous immunoglobulin reduces mortality and adverse neuro-developmental outcome.Methods and designA randomised, placebo controlled, double blind trial. Babies with suspected or proven neonatal sepsis will be randomised to receive intravenous immunoglobulin therapy or placebo.Eligibility criteriaBabies must be receiving antibiotics and have proven or suspected serious infection AND have at least one of the following: birthweight less than 1500 g OR evidence of infection in blood culture, cerebrospinal fluid or usually sterile body fluid OR be receiving respiratory support via an endotracheal tube AND there is substantial uncertainty that intravenous immunoglobulin is indicated.Exclusion criteriaBabies are excluded if intravenous immunoglobulin has already been given OR intravenous immunoglobulin is thought to be needed OR contra-indicated. Trial treatment: Babies will be given either 10 ml/kg of intravenous immunoglobulin or identical placebo solution over 4-6 hours, repeated 48 hours later.Primary outcomeMortality or major disability at two years, corrected for gestational age.Data collectionData will be collected at discharge from hospital and at 2 years of age (corrected for gestation) using a parental questionnaire and a health status questionnaire completed during a face-to-face follow-up appointment with the child's paediatrician.Trial registrationCurrent Controlled Trials ISCRTN94984750.

Project description:ObjectiveTo evaluate the efficacy of financial incentives dependent on continuous smoking abstinence on smoking cessation and birth outcomes among pregnant smokers.DesignSingle blind, randomised controlled trial.SettingFinancial Incentive for Smoking Cessation in Pregnancy (FISCP) trial in 18 maternity wards in France.Participants460 pregnant smokers aged at least 18 years who smoked ≤5 cigarettes/day or ≤3 roll-your-own cigarettes/day and had a pregnancy gestation of <18 weeks were randomised to a financial incentives group (n=231) or a control group (n=229).InterventionsParticipants in the financial incentives group received a voucher equivalent to €20 (£17; $23), and further progressively increasing vouchers at each study visit if they remained abstinent. Participants in the control group received no financial incentive for abstinence. All participants received a €20 show-up fee at each of six visits.Main outcome measuresThe main outcome measure was continuous smoking abstinence from the first post-quit date visit to visit 6, before delivery. Secondary outcomes in the mothers were point prevalence abstinence, time to smoking relapse, withdrawal symptoms, blood pressure, and alcohol and cannabis use in past 30 days. Secondary outcomes in the babies were gestational age at birth, birth characteristics (birth weight, length, head circumference, Apgar score), and a poor neonatal outcome-a composite measure of transfer to the neonatal unit, congenital malformation, convulsions, or perinatal death.ResultsMean age was 29 years. In the financial incentives and control groups, respectively, 137 (59%) and 148 (65%) were employed, 163 (71%) and 171 (75%) were in a relationship, and 41 (18%) and 31 (13%) were married. The participants had smoked a median of 60 cigarettes in the past seven days. The continuous abstinence rate was significantly higher in the financial incentives group (16%, 38/231) than control group (7%, 17/229): odds ratio 2.45 (95% confidence interval 1.34 to 4.49), P=0.004). The point prevalence abstinence rate was higher (4.61, 1.41 to 15.01, P=0.011), the median time to relapse was longer (visit 5 (interquartile range 3-6) and visit 4 (3-6), P<0.001)), and craving for tobacco was lower (β=-1.81, 95% confidence interval -3.55 to -0.08, P=0.04) in the financial incentives group than control group. Financial incentives were associated with a 7% reduction in the risk of a poor neonatal outcome: 4 babies (2%) in the financial incentives group and 18 babies (9%) in the control group: mean difference 14 (95% confidence interval 5 to 23), P=0.003. Post hoc analyses suggested that more babies in the financial incentives group had birth weights ≥2500 g than in the control group: unadjusted odds ratio 1.95 (95% confidence interval 0.99 to 3.85), P=0.055; sex adjusted odds ratio 2.05 (1.03 to 4.10), P=0.041; and sex and prematurity adjusted odds ratio 2.06 (0.90 to 4.71), P=0.086. As these are post hoc analyses, the results should be interpreted with caution.ConclusionsFinancial incentives to reward smoking abstinence compared with no financial incentives were associated with an increased abstinence rate in pregnant smokers. Financial incentives dependent on smoking abstinence could be implemented as a safe and effective intervention to help pregnant smokers quit smoking.Trial registrationClinicalTrials.gov NCT02606227.

Dataset Information

A machine-learning algorithm for neonatal seizure recognition: a multicentre, randomised, controlled trial.

Publications

A machine-learning algorithm for neonatal seizure recognition: a multicentre, randomised, controlled trial.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets