Background: Gestational diabetes mellitus (GDM) has serious effects on both mother and child. Like Type 2 Diabetes Mellitus, it is increasing in prevalence world-wide. In addition to obesity, sleep duration has been named an important risk factor. Using a large cohort study, including data from 48,787 participants of the Japan Environment and Children’s Study (JECS), we examined the association between sleep duration and both random blood glucose levels and GDM rates during pregnancy.Methods: Random blood glucose levels were measured during pregnancy. GDM diagnosis was based on the results of 75 g oral glucose tolerance test. Additional anthropometric data was collected from questionnaires for statistical analysis.Results: Compared to mothers averaging 7 to < 10 h sleep (reference group), women receiving < 5 h or ≥ 10 h sleep exhibited significantly elevated random blood glucose levels. This was associated with an elevated risk for positive GDM screening (< 5 h sleep: OR 1.17 (0.96–1.44)p = 0.126; ≥10 h sleep: OR 1.13 (1.03–1.25)p = 0.006). Calculating the risk for GDM, women sleeping < 5 h or ≥ 10 h exhibited elevated risks of 1.31-fold and 1.21 respectively. However, this trend was not found to be significant.Conclusions: Sleep is a critical factor in glucose metabolism, with both abnormally long and short sleep duration increasing random blood glucose levels in pregnant women. Moreover, the risk for positive GDM screening increases significantly with elevated sleep, ≥10 h per night. These findings are promising because they support the idea that sleep duration is a modifiable risk factor, and can be focused upon to improve health and pregnancy outcome.

This study aimed to investigate the association of maternal sleep before and during pregnancy with preterm birth, infant sleep and temperament at 1 month of age. We used the data of the Japan Environment and Children’s Study, a cohort study in Japan, which registered 103,099 pregnancies between 2011 and 2014. Participants were asked about their sleep before and during pregnancy, and the sleep and temperament of their newborns at 1 month of age. Preterm birth data were collected from medical records. Maternal sleep was not associated with preterm birth, but subjective sleep quality during pregnancy was associated with late preterm birth (birth at 34–36 weeks of gestation). For example, participants with extremely light subjective depth of sleep were more likely to experience preterm birth (RR = 1.19; 95% confidence interval [CI] = 1.04–1.35). Maternal sleep both before and during pregnancy seemed to be associated with infant sleep and temperament at 1 month of age. Infants, whose mothers slept for less than 6 hours before pregnancy, tended to cry intensely (RR = 1.15; 95% CI = 1.09–1.20). Maternal sleep problems before and during pregnancy were associated with preterm birth and child sleep problems and temperament.

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.

Evidence supporting an inverse association between maternal exposure to air pollutants and foetal growth has been accumulating. However, the findings from Asian populations are limited, and the question of critical windows of exposure remains unanswered. We examined whether maternal exposure to air pollutants, in particular exposure during the first trimester (an important period of placental development), was associated with foetal growth in Japanese term infants. From the Japan Perinatal Registry Network database, we received birth data for 29,177 term singleton births in western Japan (Kyushu-Okinawa Districts) between 2005 and 2010. Exposure was expressed in terms of average concentrations of air pollutants (ozone, suspended particulate matter, nitrogen dioxide, and sulphur dioxide), as measured at the nearest monitoring stations to the respective delivery hospitals of the pregnant women, during the entire pregnancy and each trimester. As proxy markers of foetal growth restriction, we used small for gestational age (SGA), and adverse birth weight (low birth weight in addition to SGA). For pollutant exposure during the entire pregnancy, we did not observe the association with SGA and adverse birth weight. In the single-trimester model for the first trimester, however, we found a positive association between ozone exposure, and SGA (odds ratio [OR] per 10 ppb increase = 1.07, 95% confidence interval [CI] = 1.01-1.12) and adverse birth weight (OR = 1.07; 95% CI = 1.01-1.14). This association persisted in the multi-trimester model, and no association for exposure during the second or third trimester was observed. Exposure to other pollutants during each trimester was not associated with these outcomes. In conclusion, maternal exposure to ozone during the first trimester was independently associated with an elevated risk of poor foetal growth.