Neonatal mortality is one of the key impediments in achieving global sustainable development goals, especially in lower middle income countries (LMICs). As an LMIC with the highest reported neonatal mortality rate in Southeast Asia, Indonesia faces inequitable distribution of health facilities across the archipelago. Therefore, in this paper, we aim to evaluate the determinants of neonatal mortality rate in Indonesia to search for better strategies to overcome this problem. We conducted an analysis of the 2017 Indonesia Demographic Health Survey dataset of 10,838 live-born infants born from singleton pregnancies in 2017. Using a hierarchical approach, multivariate analysis was conducted to identify potential factors (including socioeconomic, household, and proximate determinants) that contributed to neonatal mortality. The lack of participation in postnatal care [odds ratio (OR) = 20.394, p = 0.01)] and delivery complications other than prolonged labour (OR = 2.072, p = 0.02) were the maternal factors that significantly associated with increased risk of neonatal death. Regarding neonatal factors, low-birth-weight infants appeared to be more vulnerable to neonatal death (OR = 12.489, p = 0.01). Low participation in postnatal care, development of labour complications, and low birth weight were associated with higher neonatal mortality. It implies that in a limited resource and geographically challenging country such as Indonesia, improving the quality and optimizing services of public hospitals with equitable distribution of quality health care services in all regions should be prioritized in the efforts of reducing neonatal mortality rate.

Severe hyperbilirubinemia, which may result in kernicterus, is seen more frequently in low and middle-income countries, such as Indonesia, than in high-income countries. In Indonesia midwives, general practitioners (GPs), and pediatricians are involved in the care of jaundiced newborn infants. It is unknown whether the high incidence of severe hyperbilirubinemia in this country is related to a lack of awareness of existing hyperbilirubinemia guidelines issued by, for example, the World Health Organization, the American Academy of Pediatrics, or the Indonesian Health Ministry, or to a lack of adherence to such guidelines. The aim of this questionnaire study was to assess health professionals’ awareness of existing guidelines and their adherence to these guidelines in daily practice. We handed out a ten-question questionnaire to midwives, GPs, and pediatricians that included questions about the professionals themselves as well as clinical questions. The midwives completed 291 questionnaires, the GPs 206, and the pediatricians 154, all of which we used for our analysis. Almost 30% of the midwives and 23% of the GPs were either unaware of any existing guidelines or they did not adhere to them. Only 54% of the midwives recognized the warning signs of severe hyperbilirubinemia correctly, compared to 68% of the GPs and 89% of the pediatricians. Twenty-eight percent of the midwives and 31% of the GPs indicated that their first follow-up visit was after 72 hours, while 90% of them discharged infants after less than 48 hours after birth. The awareness of and adherence to guidelines for preventing and treating hyperbilirubinemia is low amongst the midwives and GPs in Indonesia. This may be an important contributing factor in the high incidence of severe hyperbilirubinemia in Indonesia.

In Indonesia, the burden of severe hyperbilirubinemia is higher compared to other countries. Whether this is related to ineffective phototherapy (PT) is unknown. The aim of this study is to investigate the performance of phototherapy devices in hospitals on Java, Indonesia. In 17 hospitals we measured 77 combinations of 20 different phototherapy devices, with and without curtains drawn around the incubator/crib. With a model to mimic the silhouette of an infant, we measured the irradiance levels with an Ohmeda BiliBlanket Meter II, recorded the distance between device and model, and compared these to manufacturers’ specifications. In nine hospitals the irradiance levels were less than required for standard PT: < 10 μW/cm2/nm and in eight hospitals irradiance failed to reach the levels for intensive phototherapy: 30 μW/cm2/nm. Three hospitals provided very high irradiance levels: > 50 μW/cm2/nm. Half of the distances between device and model were greater than recommended. Distance was inversely correlated with irradiance levels (R2 = 0.1838; P < 0.05). The effect of curtains on irradiance levels was highly variable, ranging from − 6.15 to + 15.4 μW/cm2/nm, with a mean difference (SD) of 1.82 (3.81) μW/cm2/nm (P = 0.486). In half of the hospitals that we studied on Java the levels of irradiance are too low and, in some cases, too high. Given the risks of insufficient phototherapy or adverse effects, we recommend that manufacturers provide radiometers so hospitals can optimize the performance of their phototherapy devices.

Background: Babies cannot communicate their pain properly. Several pain scores are developed, but they are subjective and have high variability inter-observer agreement. The aim of this study was to construct models that use both facial expression and infant voice in classifying pain levels and cry detection. Methods: The study included a total of 23 infants below 12-months who were treated at Dr Soetomo General Hospital. The the Face Leg Activity Cry and Consolability (FLACC) pain scale and recordings of the baby's cries were taken in the video format. A machine-learning-based system was created to detect infant cries and pain levels. Spectrograms with the Short-Time Fourier Transform were used to convert the audio data into a time-frequency representation. Facial features combined with voice features extracted by using the Deep Learning Autoencoders was used for the classification of infant pain levels. Two types of autoencoders: Convolutional Autoencoder and Variational Autoencoder were used for both faces and voices. Result: The goal of the autoencoder was to produce a latent-vector with much smaller dimensions that was still able to recreate the data with minor losses. From the latent-vectors, a multimodal data representation for Convolutional Neural Network (CNN) was used for producing a relatively high F1 score, higher than single data modal such as the voice or facial expressions alone. Two major parts of the experiment were: 1. Building the three autoencoder models, which were autoencoder for the infant’s face, amplitude spectrogram, and dB-scaled spectrogram of infant’s voices. 2. Utilising the latent-vector result from the autoencoders to build the cry detection and pain classification models. Conclusion: In this paper, four pain classifier models with a relatively good F1 score was developed. These models were combined by using ensemble methods to improve performance, which resulted in a better F1 score.

Background: Circulating into foetal circulation across the placental barrier, abnormal maternal serum lipids predispose neonates to metabolic dysfunction and thereafter affect the steroid metabolism and functions of extra-embryonic foetal tissues. Methods: A systematic review was conducted by searching PubMed–MEDLINE and the Cochrane library between January 2010 and January 2020. The included studies were English case control studies that described original data on at least one raw lipid measurement during pregnancy in healthy women who delivered large for gestational age (LGA) newborns and in healthy women with non-LGA newborns. The data extracted from 12 studies were pooled, and the weighted mean difference (WMD) in lipid levels was calculated using random effects models. A meta-analysis was performed to identify sources of heterogeneity and to describe the significant value of the collected studies. Results: Of 649 published articles identified, a total of 12 met the inclusion criteria. Compared with women who had non-LGA newborns, those who had LGA newborns had significantly higher triglyceride (TG) levels (WMD = 0.28, 95% CI −0.02 to 0.54) and lower high density lipoprotein cholestrol (HDL-C) levels (WMD = 0.08, 95% CI −0.13 to −0.03), but not have significantly lower high-density lipoprotein cholesterol (LDL-C) levels. Moreover, the levels of total cholesterol, low-density lipoprotein cholesterol, and very low density lipoprotein cholesterol (VLDL-C) were inconsistent between both groups. Conclusions: High levels of TG and low levels of HDL-C could cause births of LGA newborns whereas maternal serum of TC, LDL-C and VLDL-C cannot be used as predictor of LGA.