Sort by
Machine Learning to Improve Accuracy of Transcutaneous Bilirubinometry

Plain Language SummaryNeonatal jaundice is a common condition that might occur in the first week of life. Severe high bilirubin levels can lead to long-term problems like cerebral palsy, hearing impairment, and developmental delay. Therefore, newborns are usually evaluated for jaundice every day during their first week. To accurately assess jaundice, we need to measure the total bilirubin in the blood. However, daily blood draws for bilirubin measurements can be uncomfortable and may cause anemia in newborns. Thus, transcutaneous bilirubinometry is devised. It is a device that measures bilirubin through the skin and is commonly used because it is non-invasive. However, transcutaneous bilirubinometry has some accuracy issues. In this study, we successfully improved the accuracy of bilirubin measurement by combining machine learning with transcutaneous bilirubinometry. Being used alone, the transcutaneous bilirubinometer had an error of 1.08 mg/dL. However, combining transcutaneous bilirubinometry with machine learning, the error decreased to 0.80 mg/dL, which is a 25% of improvement. Using this approach, unnecessary blood draws could be reduced by up to 78%. If we incorporate this algorithm into transcutaneous bilirubinometry, this novel method has the potential to improve prediction accuracy and reduce the burden on babies.

Just Published
Relevant
Severe Bronchopulmonary Dysplasia Adversely Affects Brain Growth in Preterm Infants

Introduction: Bronchopulmonary dysplasia (BPD) is associated with neurodevelopmental outcomes of preterm infants, but its effect on brain growth in preterm infants after the neonatal period is unknown. This study aimed to evaluate the effect of severe BPD on brain growth of preterm infants from term to 18 months of corrected age (CA). Methods: Sixty-three preterm infants (42 with severe BPD and 21 without severe BPD) who underwent magnetic resonance imaging at term equivalent age (TEA) and 18 months of CA were studied by using the Infant Brain Extraction and Analysis Toolbox (iBEAT). We measured segmented brain volumes and compared brain volume and brain growth velocity between the severe BPD group and the non-severe BPD group. Results: There was no significant difference in brain volumes at TEA between the groups. However, the brain volumes of the total brain and cerebral white matter in the severe BPD group were significantly smaller than those in the non-severe BPD group at 18 months of CA. The brain growth velocities from TEA to 18 months of CA in the total brain, cerebral cortex, and cerebral white matter in the severe BPD group were lower than those in the non-severe BPD group. Conclusion: Brain growth in preterm infants with severe BPD from TEA age to 18 months of CA is less than that in preterm infants without severe BPD.

Relevant
Relationship between Neonatal Cerebral Fuels and Neurosensory Outcomes at 3 Years in Well Babies: Follow-Up of the Glucose in Well Babies (GLOW) Study

Plain Language SummaryIn the Glucose in Well babies and their later Neurodevelopment (GLOWiNg) study, we investigated whether the availability of brain fuels in healthy term babies during the first 5 days after birth was associated with their neurodevelopmental progress at 3 years of age. We had measured plasma glucose, lactate, and beta-hydroxybutyrate concentrations when the children were babies, as these are the most common brain fuels. We used the Ages and Stages questionnaire, which is completed by parents, to assess the child’s development in areas including communication, motor skills (fine and gross), and problem solving.We studied 64 families whose babies were part of the Glucose in Well Babies (GLOW) study in New Zealand. Using both structural equation modelling and linear regression, we found that even though many babies experienced episodes of low plasma glucose concentrations, there was no detectable relationship between these episodes and developmental progress at 3 years of age. We explored other factors, including glucose concentrations, total fuels available, and the duration of low glucose episodes, and found no significant associations. Our results suggest that in healthy term babies, episodes of low glucose concentrations in the first days after birth may be part of the normal metabolic transition after birth and may not have long-term implications for neurodevelopment. However, larger studies using different assessment tools are needed to confirm these findings.

Relevant
Mortality Risk in US Neonatal Intensive Care Unit Infants by Birth Size Classifications Comparing Three Growth Curves

Introduction: Three widely referenced growth curves classify infant birth anthropometric measurements as small (SGA), appropriate (AGA), or large (LGA) for gestational age (GA) differently. We assessed how these differences in assignment affect the identification and prediction of neonatal intensive care unit (NICU) mortality risk in US preterm infants. Methods: Birth data of infants admitted to NICUs from the Pediatrix Clinical Data Warehouse (2013–2018) were analyzed. Birth weight, length, and head circumference of 46,724 singleton infants (24–32 weeks GA) were classified as SGA, AGA, or LGA using the Olsen, Fenton, and INTERGROWTH-21st curves. NICU mortality risk based on birth size classification was analyzed using unadjusted and adjusted logistic regression stratified by GA. Results: Odds of mortality were increased with SGA classification at all GAs, size measurements, and curve sets, compared with AGA infants. LGA classification for weight was associated with lower mortality risk at 24 weeks GA and higher risk at 30 weeks GA. Odds of mortality did not differ significantly across curve sets. Classification of size at birth alone had relatively low predictive ability to identify mortality risk, with unadjusted AUCs near 0.5 for all analyses. Conclusion: There were no significant differences across curve sets in predicting mortality. Classification of size at birth is a relatively imprecise method to identify infants at risk for NICU mortality.

Relevant
Proactive Diagnosis and Tailor-Made Treatment of Patent Ductus Arteriosus in Very Preterm Infants with Routine Echocardiography in Japan: A post hoc Analysis of the PLASE Study

Introduction: A feature of the management of extremely preterm infants in Japan is proactive circulatory management using early routine echocardiography performed by neonatologists. Methods: This study was a post hoc analysis of the Patent ductus arteriosus and Left Atrial Size Evaluation in preterm infants (PLASE) study, which is a prospective cohort study including preterm infants admitted to 34 tertiary neonatal intensive care units in Japan between October 2015 and December 2016. We described the details of the treatment strategy of patent ductus arteriosus (PDA) based on early routine echocardiography. Results: In total, 613 preterm infants were included into the analysis. Twenty percent of infants with prophylactic indomethacin were switched to therapeutic cyclooxygenase inhibitor (COX-I) before the completion of the full prophylactic indomethacin course. Therapeutic COX-I was mostly administered based on echocardiographic findings before PDA became symptomatic or hemodynamically significant. Therapeutic COX-I was frequently discontinued after one or two doses before the full course (three doses) was completed. The proportion of infants requiring additional treatment (additional therapeutic COX-I course or surgical PDA closure) after discontinued COX-I courses (<3 doses) compared to infants after completed 3 doses course was significantly lower (after the first therapeutic COX-I course 46% vs. 68%, p < 0.001) or without a significant difference (after the second or third course). Conclusions: The clinical management of PDA in Japan featured (1) COX-I administration based on echocardiographic findings before symptomatic or hemodynamically significant PDA appeared and (2) frequent discontinuation of therapeutic COX-I before completing the standard three doses course.

Relevant
Coupling between Regional Oxygen Saturation of the Brain and Vital Signs during Immediate Transition after Birth

Introduction: The primary aim was to analyze any coupling of heart rate (HR)/arterial oxygen saturation (SpO<sub>2</sub>) and regional cerebral oxygen saturation (rScO<sub>2</sub>) and regional cerebral fractional tissue oxygen extraction (cFTOE) during immediate transition after birth in term and preterm neonates to gain more insight into interactions. Methods: The present study is a post hoc analysis of data from 106 neonates, obtained from a prospective, observational study. Measurements of HR, SpO<sub>2</sub>, rScO<sub>2</sub>, and cFTOE were performed during the first 15 min after birth. The linear and nonlinear correlation were computed between these parameters in a sliding window. The resulting coupling curves were clustered. After clustering, demographic data of the clusters were de-blinded and compared. Results: Due to missing data, 58 out of 106 eligible patients were excluded. Two clusters were obtained: cluster 1 (N = 39) and cluster 2 (N = 9). SpO<sub>2</sub> had linear and nonlinear correlations with rScO<sub>2</sub> and cFTOE, whereby the correlations with rScO<sub>2</sub> were more pronounced in cluster 2. HR-rScO<sub>2</sub> and HR-cFTOE demonstrated a nonlinear correlation in both clusters, again being more pronounced in cluster 2, whereby linear correlations were mainly absent. After de-blinding, the demographic data revealed that the neonates in cluster 2 had significantly lower gestational age (mainly preterm) compared to cluster 1 (mainly term). Discussion: Besides SpO<sub>2</sub>, also HR demonstrated a nonlinear correlation with rScO<sub>2</sub> and cFTOE in term and preterm neonates during immediate transition after birth. In addition, the coupling of SpO<sub>2</sub> and HR with cerebral oxygenation was more pronounced in neonates with a lower gestational age.

Relevant
Endothelial-to-Mesenchymal Transition in Human and Murine Models of Congenital Diaphragmatic Hernia

Introduction: Congenital diaphragmatic hernia (CDH) is a complex congenital disorder, characterized by pulmonary hypertension (PH) and hypoplasia. PH secondary to CDH (CDH-PH) features devastating morbidity and mortality (25–30%) among neonates. An unmet need is determining mechanisms triggering CDH-PH to save infants. Prior data suggest abnormal remodeling of the pulmonary vascular extracellular matrix (ECM), presumed to be driven by endothelial-to-mesenchymal transition (EndoMT), hinders postnatal vasodilation and limits anti-PH therapy in CDH. There are limited data on the role of EndoMT in CDH-PH. Methods: The purpose of the study was to investigate how EndoMT contributes to CDH-PH by identifying cells undergoing EndoMT noted by alpha smooth muscle actin (α-SMA) expression in human umbilical vein endothelial cells (HUVECs) and lung tissue obtained from murine pups using the nitrofen model. N = 8 CDH, N = 8 control HUVECs were stained for α-SMA and CD31 after being exposed for 24 h to TGFB, a known EndoMT promoter. N = 8 nitrofen, N = 8 control murine pup lungs were also stained for α-SMA and CD31. α-SMA and CD31 expression was quantified in HUVECs and murine tissue using Fiji imaging software and normalized to the total number of cells per slide noted by DAPI staining. Results: CDH HUVECs demonstrated a 1.1-fold increase in α-SMA expression (p = 0.02). The murine model did not show statistical significance between nitrofen and control pup lungs; however, there was a 0.4-fold increase in α-SMA expression with a 0.8-fold decrease in CD31 expression in the nitrofen pup lungs when compared to controls. Conclusion: These results suggest that EndoMT could potentially play a role in the ECM remodeling seen in CDH-PH.

Relevant
Long-Term Growth and Neurodevelopmental Outcomes of Neonates Infected with SARS-CoV-2 during the COVID-19 Pandemic at 18–24 Months Corrected Age: A Prospective Observational Study

Introduction: During the early coronavirus disease (COVID-19) pandemic in 2020, researchers cautioned about the potential neuroinvasive capability of the virus and long-term neurological consequences. Although a few preliminary studies have found delayed communication, fine motor, and problem-solving skills in infants after COVID-19 infection, there continues to be a paucity of data on long-term development of neonates diagnosed with COVID-19. Methods: We conducted a prospective study of 20 neonates who acquired severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection during the first wave of the pandemic (April–July 2020). At 18–24 months corrected age, we assessed neurodevelopment by Bayley Scales of Infant and Toddler Development, the third edition (BSID-III), along with growth, hearing, and vision evaluation. Results: The mean corrected age at assessment was 21 months 11 days ± 1 month 28 days. We found developmental delay in nearly half of the children with scores below one standard deviation in either of the BSID-III domains. Mild delay in either motor, cognitive, or language domains was found in 9 (45%) children and moderate delay in 2 (10%). Expressive language, fine motor, and receptive language were predominantly affected. None of the children had hearing impairment, blindness, or significant growth faltering including clinically severe microcephaly. The mean composite cognitive, language, and motor scores were significantly lower in those with neurodevelopmental delay (p value – 0.02, 0.000, and 0.03, respectively) without any differences in their disease characteristics. Conclusion: Neonates infected with SARS-CoV-2 have an increased risk of developmental delays in expressive language, fine motor, and receptive language skills at 18–24 months of age. The severity of delays is predominantly mild.

Open Access
Relevant
Computer Vision for Identification of Increased Fetal Heart Variability in Cardiotocogram

Introduction: Increased fetal heart rate variability (IFHRV), defined as fetal heart rate (FHR) baseline amplitude changes of >25 beats per minute with a duration of ≥1 min, is an early sign of intrapartum fetal hypoxia. This study evaluated the level of agreement of machine learning (ML) algorithms-based recognition of IFHRV patterns with expert analysis. Methods: Cardiotocographic recordings and cardiotocograms from 4,988 singleton term childbirths were evaluated independently by two expert obstetricians blinded to the outcomes. Continuous FHR monitoring with computer vision analysis was compared with visual analysis by the expert obstetricians. FHR signals were graphically processed and measured by the computer vision model labeled SALKA. Results: In visual analysis, IFHRV pattern occurred in 582 cardiotocograms (11.7%). Compared with visual analysis, SALKA recognized IFHRV patterns with an average Cohen’s kappa coefficient of 0.981 (95% CI: 0.972–0.993). The sensitivity of SALKA was 0.981, the positive predictive rate was 0.822 (95% CI: 0.774–0.903), and the false-negative rate was 0.01 (95% CI: 0.00–0.02). The agreement between visual analysis and SALKA in identification of IFHRV was almost perfect (0.993) in cases (N = 146) with neonatal acidemia (i.e., umbilical artery pH <7.10). Conclusions: Computer vision analysis by SALKA is a novel ML technique that, with high sensitivity and specificity, identifies IFHRV features in intrapartum cardiotocograms. SALKA recognizes potential early signs of fetal distress close to those of expert obstetricians, particularly in cases of neonatal acidemia.

Relevant