Machine learning prediction of gestational age from metabolic screening markers resistant to ambient temperature transportation: Facilitating use of this technology in low resource settings of South Asia and East Africa.

Background: Knowledge of gestational age is critical for guiding preterm neonatal care. In the last decade, metabolic gestational dating approaches emerged in response to a global health need; because in most of the developing world, accurate antenatal gestational age estimates are not feasible. These methods initially developed in North America have now been externally validated in two studies in developing countries, however, require shipment of samples at sub-zero temperature. Methods: A subset of 330 pairs of heel prick dried blood spot samples were shipped on dry ice and in ambient temperature from field sites in Tanzania, Bangladesh and Pakistan to laboratory in Iowa (USA). We evaluated impact on recovery of analytes of shipment temperature, developed and evaluated models for predicting gestational age using a limited set of metabolic screening analytes after excluding 17 analytes that were impacted by shipment conditions of a total of 44 analytes. Results: With the machine learning model using all the analytes, samples shipped in dry ice yielded a Root Mean Square Error (RMSE) of 1.19 weeks compared to 1.58 weeks for samples shipped in ambient temperature. Out of the 44 screening analytes, recovery of 17 analytes was significantly different between the two shipment methods and these were excluded from further machine learning model development. The final model, restricted to stable analytes provided a RMSE of 1.24 (95% confidence interval (CI) = 1.10-1.37) weeks for samples shipped on dry ice and RMSE of 1.28 (95% CI = 1.15-1.39) for samples shipped at ambient temperature. Analysis for discriminating preterm births (gestational age <37 weeks), yielded an area under curve (AUC) of 0.76 (95% CI = 0.71-0.81) for samples shipped on dry ice and AUC of 0.73 (95% CI = 0.67-0.78) for samples shipped in ambient temperature. Conclusions: In this study, we demonstrate that machine learning algorithms developed using a sub-set of newborn screening analytes which are not sensitive to shipment at ambient temperature, can accurately provide estimates of gestational age comparable to those from published regression models from North America using all analytes. If validated in larger samples especially with more newborns <34 weeks, this technology could substantially facilitate implementation in LMICs.

Burden and risk factors for antenatal depression and its effect on preterm birth in South Asia: A population-based cohort study.

INTRODUCTION: Women experience high rates of depression, particularly during pregnancy and the postpartum periods. Using population-based data from Bangladesh and Pakistan, we estimated the burden of antenatal depression, its risk factors, and its effect on preterm birth. METHODS: The study uses the following data: maternal depression measured between 24 and 28 weeks of gestation using the 9-question Patient Health Questionnaire (PHQ-9); data on pregnancy including an ultrasound before 19 weeks of gestation; data on pregnancy outcomes; and data on woman's age, education, parity, weight, height, history of previous illness, prior miscarriage, stillbirth, husband's education, and household socioeconomic data collected during early pregnancy. Using PHQ-9 cutoff score of ≥12, women were categorized into none to mild depression or moderate to moderately severe depression. Using ultrasound data, preterm birth was defined as babies born <37 weeks of gestation. To identify risk ratios (RR) for antenatal depression, unadjusted and adjusted RR and 95% confidence intervals (CI) were calculated using log- binomial model. Log-binomial models were also used for determining the effect of antenatal depression on preterm birth adjusting for potential confounders. Data were analyzed using Stata version 16 (StataCorp LP). RESULTS: About 6% of the women reported moderate to moderately severe depressive symptoms during the antenatal period. A parity of ≥2 and the highest household wealth status were associated with an increased risk of depression. The overall incidence of preterm birth was 13.4%. Maternal antenatal depression was significantly associated with the risk of preterm birth (ARR, 95% CI: 1.34, 1.02-1.74). CONCLUSION: The increased risk of preterm birth in women with antenatal depression in conjunction with other significant risk factors suggests that depression likely occurs within a constellation of other risk factors. Thus, to effectively address the burden of preterm birth, programs require developing and providing integrated care addressing multiple risk factors.

Machine learning guided postnatal gestational age assessment using new-born screening metabolomic data in South Asia and sub-Saharan Africa.

BACKGROUND: Babies born early and/or small for gestational age in Low and Middle-income countries (LMICs) contribute substantially to global neonatal and infant mortality. Tracking this metric is critical at a population level for informed policy, advocacy, resources allocation and program evaluation and at an individual level for targeted care. Early prenatal ultrasound examination is not available in these settings, gestational age (GA) is estimated using new-born assessment, last menstrual period (LMP) recalls and birth weight, which are unreliable. Algorithms in developed settings, using metabolic screen data, provided GA estimates within 1-2 weeks of ultrasonography-based GA. We sought to leverage machine learning algorithms to improve accuracy and applicability of this approach to LMICs settings. METHODS: This study uses data from AMANHI-ACT, a prospective pregnancy cohorts in Asia and Africa where early pregnancy ultrasonography estimated GA and birth weight are available and metabolite screening data in a subset of 1318 new-borns were also available. We utilized this opportunity to develop machine learning (ML) algorithms. Random Forest Regressor was used where data was randomly split into model-building and model-testing dataset. Mean absolute error (MAE) and root mean square error (RMSE) were used to evaluate performance. Bootstrap procedures were used to estimate confidence intervals (CI) for RMSE and MAE. For pre-term birth identification ROC analysis with bootstrap and exact estimation of CI for area under curve (AUC) were performed. RESULTS: Overall model estimated GA had MAE of 5.2 days (95% CI 4.6-6.8), which was similar to performance in SGA, MAE 5.3 days (95% CI 4.6-6.2). GA was correctly estimated to within 1 week for 85.21% (95% CI 72.31-94.65). For preterm birth classification, AUC in ROC analysis was 98.1% (95% CI 96.0-99.0; p < 0.001). This model performed better than Iowa regression, AUC Difference 14.4% (95% CI 5-23.7; p = 0.002). CONCLUSIONS: Machine learning algorithms and models applied to metabolomic gestational age dating offer a ladder of opportunity for providing accurate population-level gestational age estimates in LMICs settings. These findings also point to an opportunity for investigation of region-specific models, more focused feasible analyte models, and broad untargeted metabolome investigation.

Using AMANHI-ACT cohorts for external validation of Iowa new-born metabolic profiles based models for postnatal gestational age estimation.

BACKGROUND: Globally, 15 million infants are born preterm and another 23.2 million infants are born small for gestational age (SGA). Determining burden of preterm and SGA births, is essential for effective planning, modification of health policies and targeting interventions for reducing these outcomes for which accurate estimation of gestational age (GA) is crucial. Early pregnancy ultrasound measurements, last menstrual period and post-natal neonatal examinations have proven to be not feasible or inaccurate. Proposed algorithms for GA estimation in western populations, based on routine new-born screening, though promising, lack validation in developing country settings. We evaluated the hypothesis that models developed in USA, also predicted GA in cohorts of South Asia (575) and Sub-Saharan Africa (736) with same precision. METHODS: Dried heel prick blood spots collected 24-72 hours after birth from 1311 new-borns, were analysed for standard metabolic screen. Regression algorithm based, GA estimates were computed from metabolic data and compared to first trimester ultrasound validated, GA estimates (gold standard). RESULTS: Overall Algorithm (metabolites + birthweight) estimated GA to within an average deviation of 1.5 weeks. The estimated GA was within the gold standard estimate by 1 and 2 weeks for 70.5% and 90.1% new-borns respectively. Inclusion of birthweight in the metabolites model improved discriminatory ability of this method, and showed promise in identifying preterm births. Receiver operating characteristic (ROC) curve analysis estimated an area under curve of 0.86 (conservative bootstrap 95% confidence interval (CI) = 0.83 to 0.89); P < 0.001) and Youden Index of 0.58 (95% CI = 0.51 to 0.64) with a corresponding sensitivity of 80.7% and specificity of 77.6%. CONCLUSION: Metabolic gestational age dating offers a novel means for accurate population-level gestational age estimates in LMIC settings and help preterm birth surveillance initiatives. Further research should focus on use of machine learning and newer analytic methods broader than conventional metabolic screen analytes, enabling incorporation of region-specific analytes and cord blood metabolic profiles models predicting gestational age accurately.