We developed a deep learning–based extraction of electrocardiographic (ECG) waves from ballistocardiographic (BCG) signals and explored their use in R–R interval (RRI) estimation. Preprocessed BCG and reference ECG signals were inputted into the bidirectional long short-term memory network to train the model to minimize the loss function of the mean squared error between the predicted ECG (pECG) and genuine ECG signals. Using a dataset acquired with polyvinylidene fluoride and ECG sensors in different recumbent positions from 18 participants, we generated pECG signals from preprocessed BCG signals using the learned model and evaluated the RRI estimation performance by comparing the predicted RRI with the reference RRI obtained from the ECG signal using a leave-one-subject-out cross-validation scheme. A mean absolute error (MAE) of 0.034 s was achieved for the beat-to-beat interval accuracy. To further test the generalization ability of the learned model trained with a short-term-recorded dataset, we collected long-term overnight recordings of BCG signals from 12 different participants and performed validation. The beat-to-beat interval correlation between BCG and ECG signals was 0.82 ± 0.06 with an average MAE of 0.046 s, showing practical performance for long-term measurement of RRIs. These results suggest that the proposed approach can be used for continuous heart rate monitoring in a home environment.

Portable medical sensors play an important role in healthcare services, especially in rural communities. Many telehealth systems use these devices for providing patients' vital information from a distance to remote doctors. Erroneous data will not only mislead the remote doctor for correct diagnosis but it will cause health threats to these unreached community people. Therefore, it is very important to identify good sensors with an acceptable level of accuracy but within the affordable price of the available sensors in the market. This study aims to identify quality portable cholesterol sensors with high accuracy with the reference of the Japanese clinical pathology laboratory as a gold standard. We have considered cholesterol sensors that measure total cholesterol for this study that are commonly used in the developing countries of Asia. We found that out of four, three of them were very much erroneous and cannot be recommended even for primary healthcare.

To examine whether patterns, such as the timings of onset or recovery from sleep disturbance, are associated with later developmental problems, including autism spectrum disorder (ASD). Mothers participating in the Japan Environment and Children's Study with a child aged 3 years were included in the analyses. Children were assessed for short sleep and frequent awakenings at 1 month, 6 months, and 1 year of age. Developmental problems were evaluated at 3 years of age based on ASD diagnosis and developmental delay, using the Japanese translation of the Ages and Stages Questionnaire (ASQ) 3rd edition. Sleep disturbance patterns were classified by onset age, and developmental problem risks were examined based on onset/recovery ages. Among 63,418 mother-infant dyads, 0.4% of infants were later diagnosed with ASD, and 14.4% had abnormal scores on any ASQ domains. The later the onset of short sleep, the lower the risk of abnormal ASQ scores (RR of short sleep onset at 1 year: 1.41; 6 months: 1.52; 1 month: 1.57). The earlier the infants recovered from short sleep persistence, the lower the risk of developmental delay (RR of remittance of sleep problems identified at 1 month by 6 months: 1.07; 1 year: 1.31; not before 1 year: 1.57). Although not all patterns were significant, later short sleep onset and earlier recovery were associated with lower ASD risk. These findings may have significant implications for future interventions in infant development.

In this study, we aimed to determine the association of premenstrual syndrome (PMS) or premenstrual dysphoric disorder (PMDD) with maternal antenatal depression. This cross-sectional, online questionnaire-based observational study included 212 pregnant women between gestational ages 24 weeks and 28 weeks 6 days. PMS and PMDD were measured using the PMDD Scale, and maternal antenatal depression was evaluated using the Edinburgh Postnatal Depression Scale. Baseline characteristics, clinical information, and associated factors were also included in the questionnaire. Analyses were conducted using a binomial logistic regression model with Edinburgh Postnatal Depression Scale positivity (maternal antenatal depression) as the dependent variable. Having "PMDD" (odds ratio: 3.54 [95% confidence interval: 1.26-9.93], p = 0.02) and "PMS" (odds ratio 2.31 [1.10-4.87], p = 0.03) on the PMDD rating scale were significantly associated with maternal antenatal depression. Therefore, our results suggest that screening for a history of PMS or PMDD during the early antepartum interview may aid mental health care and prevent perinatal depression during the early stages of pregnancy.

Telehealth systems in underserved countries leverage various low-cost portable medical sensors to transmit patients' vital information to remote doctors, facilitating timely diagnoses and interventions. However, the potential risks associated with inaccurate data pose considerable threats to the health of individuals. This study focuses on identifying high-quality portable hemoglobin sensors, employing the Japanese clinical pathology laboratory as a gold standard. Out of the eight sensors evaluated in this study, four were found to be highly erroneous.

This study aimed to determine the longitudinal effects of pre-pregnancy and pregnancy sleep on maternal depression at one month postpartum, stratified by parity. We used data from the Japan Environment and Children’s Study, a cohort study that registered 103,060 pregnancies between 2011 and 2014. A total of 76,977 women were included in this study and were asked about their sleep pre-pregnancy and during pregnancy, and whether they had postpartum depression. We found that those who slept 8–9 hours in pre-pregnancy had slightly lower odds of postpartum depression compared with those who slept 7–8 hours (odds ratio OR = 0.92, 95% confidence interval CI:0.87–0.99), and those who had bedtimes other than 9 p.m. to 3 a.m. had increased OR compared with those who had bedtimes between 9 p.m. and midnight (OR = 1.14, 95% CI:1.00–1.31). During pregnancy, those who slept less than 6 hours and whose bedtime was other than 9 p.m. to 3 a.m. had increased ORs of postpartum depression (OR = 1.21, 95% CI:1.09–1.31; OR = 1.25, 95% CI:1.08–1.45, respectively). Poor sleep quality during pre-pregnancy and pregnancy increased the odds of postpartum depression. To prevent postpartum depression, it is important for women to have good sleep from pre-pregnancy.

Maternal and newborn care quality can be measured in three dimensions (Dimensions 1: care provision, 2: care experience, and 3: human and physical resources); however, little is known about which dimensions are associated with newborn and perinatal deaths. We examined the association between care quality and newborn and perinatal deaths in Nepal. This study incorporated secondary data from Nepal Service Provision Assessments (NSPA) 2015 (623 delivery facilities, facility inventory survey; 1,509 women, ANC clients interviews; 1,544 women, ANC observation) and Nepal Demographic and Health Surveys (NDHS) 2016 (5,038 women who reported having given birth in the five years preceding data collection). The outcome variables were newborn and perinatal deaths derived from the NDHS. The exposure variables were district-level maternal and newborn care quality scores calculated from the NSPA data. Covariates were women's sociodemographic, health, and obstetric characteristics. We applied the administrative boundary method to link these two surveys. We conducted binary logistic regression analyses to examine the association between care quality and newborn/perinatal deaths. In Dimension 1, higher mean and maximum quality scores at the district level were associated with a lower number of newborn deaths (mean: odds ratio [OR] = 0.04, 95% confidence interval [CI]: 0.00-0.76; max: OR = 0.09, 95% CI: 0.01-0.58), but not with perinatal deaths. In Dimensions 2 and 3, the quality score was not significantly associated with newborn deaths and perinatal. Enhancing the quality of care provision at its average and highest levels in each district may contribute to the reduction of newborn deaths, but not perinatal death. Health administrators should assess the quality of care at the administrative division level and focus on enhancing both average and maximum care quality of health facilities in each region in the care provision dimension.

Deep learning methods have gained significant attention in sleep science. This study aimed to assess the performance of a deep learning-based sleep stage classification model constructed using fewer physiological parameters derived from cardiorespiratory and body movement data. Overnight polysomnography (PSG) data from 123 participants (age: 19–82 years) with suspected sleep disorders were analyzed. Multivariate time series data, including heart rate, respiratory rate, cardiorespiratory coupling, and body movement frequency, were input into a bidirectional long short-term memory (biLSTM) network model to train and predict five-class sleep stages. The trained model's performance was evaluated using balanced accuracy, Cohen's κ coefficient, and F1 scores on an epoch-per-epoch basis and compared with the ground truth using the leave-one-out cross-validation scheme. The model achieved an accuracy of 71.2 ± 5.8%, Cohen's κ of 0.425 ± 0.115, and an F1 score of 0.650 ± 0.083 across all sleep stages, and all metrics were negatively correlated with the apnea–hypopnea index, as well as age, but positively correlated with sleep efficiency. Moreover, the model performance varied for each sleep stage, with the highest F1 score observed for N2 and the lowest for N3. Regression and Bland–Altman analyses between sleep parameters of interest derived from deep learning and PSG showed substantial correlations ( r = 0.33–0.60) with low bias. The findings demonstrate the efficacy of the biLSTM deep learning model in accurately classifying sleep stages and in estimating sleep parameters for sleep structure analysis using a reduced set of physiological parameters. The current model without using EEG information may expand the application of unobtrusive in-home monitoring to clinically assess the prevalence of sleep disorders outside of a sleep laboratory.

Abstract Background Sleep problems and irritable temperaments are common among infants with autism spectrum disorder (ASD). The prospective association between such sleep problems and irritable temperaments and ASDs needs to be determined for elucidating the mechanism and exploring the future intervention study. Thus, in this study, we investigated whether sleep quality and temperament in 1-month-old infants are associated with the onset of ASD in 3-year-old children. We also assessed its sex-stratified associations. Methods We conducted a longitudinal study using data from 69,751 mothers and infants from a large-cohort study, the Japan Environment and Children’s Study. We examined the prospective association between infant sleep quality and temperament at 1 month of age and ASD diagnosis by 3 years of age. Results Here we show infants with longer daytime sleep have a higher risk of later ASD than those with shorter daytime sleep (risk ratio [RR]: 1.33, 95% confidence interval [CI]: 1.01–1.75). Infants who experienced intense crying have a higher risk of ASD than those who did not (RR: 1.31, 95% CI: 1.00–1.72). There is a difference in sex in the association between a bad mood and later ASD. In particular, female infants experiencing bad moods have a higher risk of ASD than others (RR: 3.59, 95% CI: 1.91–6.75). Conclusions The study findings provide important information for future intervention to reduce the risk of future ASD.

BACKGROUND:Ensuring an appropriate continuum of care in maternal, newborn, and child health, as well as providing nutrition care, is challenging in remote areas. To make care accessible for mothers and infants, we developed a telehealth care system called Portable Health Clinic for Maternal, Newborn, and Child Health. OBJECTIVE:Our study will examine the telehealth care system's effectiveness in improving women's and infants' care uptake and detecting their health problems. METHODS:A quasi-experimental study will be conducted in rural Bangladesh. Villages will be allocated to the intervention and control areas. Pregnant women (≥16 gestational weeks) will participate together with their infants and will be followed up 1 year after delivery or birth. The intervention will include regular health checkups via the Portable Health Clinic telehealth care system, which is equipped with a series of sensors and an information system that can triage participants' health levels based on the results of their checkups. Women and infants will receive care 4 times during the antenatal period, thrice during the postnatal period, and twice during the motherhood and childhood periods. The outcomes will be participants' health checkup coverage, gestational and neonatal complication rates, complementary feeding rates, and health-seeking behaviors. We will use a multilevel logistic regression and a generalized estimating equation to evaluate the intervention's effectiveness. RESULTS:Recruitment began in June 2020. As of June 2022, we have consented 295 mothers in the study. Data collection is expected to conclude in June 2024. CONCLUSIONS:Our new trial will show the effectiveness and extent of using a telehealth care system to ensure an appropriate continuum of care in maternal, newborn, and child health (from the antenatal period to the motherhood and childhood periods) and improve women's and infants' health status. TRIAL REGISTRATION:ISRCTN Registry ISRCTN44966621; https://www.isrctn.com/ISRCTN44966621. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID):DERR1-10.2196/41586.

Objectives This study aimed to investigate the continuum of care (CoC) completion rate in maternal, neonatal and child health and its associated factors among mothers in two ecological regions in Nepal.Design This was a community-based, cross-sectional study, for which data were collected through face-to-face interviews using a structured questionnaire. Multiple logistic regression analyses were conducted to determine the associated factors.Setting This was carried out in two rural districts of Nepal, in different regions: one in the hills (Dhading) and another in the flatlands called Terai (Nawalparasi). The data were collected between July and December 2016.Participants Mothers who gave birth within a year before this study were included as participants. In total, there were 1803 participants.An outcome measure The outcome of this study was measured by the CoC completion rate when a mother completes four antenatal check-ups, deliver at a health facility and receives postnatal care within 24 hours of delivery.Results The CoC completion rates were 41% in Dhading and 28% in Nawalparasi. In Dhading, shorter travel time to a health facility and higher wealth quintiles were associated with a better CoC completion rate. In Nawalparasi, the CoC completion rate was affected by parity and decision-making for pregnancy care.Conclusions The CoC completion rate was low in both districts in Nepal. However, factors associated with the CoC completion rate varied by district. Differences in these factors might be reflected by geographical and socioeconomic conditions and the characteristics of household decision making in these districts.

Objective To measure the length of stay at a health facility after childbirth, identify factors associated with the length of stay and measure the gap between the timings of the last check-up and discharge. Design A cross-sectional study. Setting Five public health facilities in Dhading, Nepal. Participants 351 randomly selected mothers who gave birth at selected health facilities within 1 year of data collection between 10 and 31 August 2018. Outcome measure Length of stay (hours) at a health facility after childbirth. Adequate length was defined as 24 hours or longer based on the WHO guidelines. Results Among 350 mothers (99.7%) out of 351 recruited, 61.7% were discharged within 24 hours after childbirth. Factors associated with shorter length of stay were as follows: travel time less than 30 min to a health facility (incidence rate ratio (IRR)=0.69, 95% CI 0.61 to 0.78); delivery attended by auxiliary staff (IRR=0.86, 95% CI 0.75 to 0.98); and delivery in a primary healthcare centre (IRR=0.67, 95% CI 0.58 to 0.79). Factors associated with longer length of stay were as follows: aged 22 years or above at the first pregnancy (IRR=1.25, 95% CI 1.13 to 1.40); having maternal complications (IRR=2.41, 95% CI 2.16 to 2.70); accompanied by her own family (IRR=1.17, 95% CI 1.03 to 1.34), accompanied by her husband (IRR=1.16, 95% CI 1.04 to 1.29); and delivered at a facility with a physical space where mother and newborn could stay overnight (IRR=1.20, 95% CI 1.07 to 1.34). Among mothers without complications, 32% received the last check-up 3 hours or less before discharge. Conclusions Multiple factors, such as mothers' conditions, health facility characteristics and external support, were associated with the length of stay after childbirth. However, even if mothers stayed long, they might have not necessarily received timely and proper assessment before discharge.