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.

A Novel Icterometer for Hyperbilirubinemia Screening in Low-Resource Settings.

BACKGROUND: Severe neonatal hyperbilirubinemia (>20 mg/dL) affects ∼1 million infants annually. Improved jaundice screening in low-income countries is needed to prevent bilirubin encephalopathy and mortality. METHODS: The Bili-ruler is an icterometer for the assessment of neonatal jaundice that was designed by using advanced digital color processing. A total of 790 newborns were enrolled in a validation study at Brigham and Women's Hospital (Boston) and Sylhet Osmani Medical College Hospital (Sylhet, Bangladesh). Independent Bili-ruler measurements were made and compared with reference standard transcutaneous bilirubin (TcB) and total serum bilirubin (TSB) concentrations. RESULTS: Bili-ruler scores on the nose were correlated with TcB and TSB levels (r = 0.76 and 0.78, respectively). The Bili-ruler distinguished different clinical thresholds of hyperbilirubinemia, defined by TcB, with high sensitivity and specificity (score ≥3.5: 90.1% [95% confidence interval (CI): 84.8%-95.4%] and 85.9% [95% CI: 83.2%-88.6%], respectively, for TcB ≥13 mg/dL). The Bili-ruler also performed reasonably well compared to TSB (score ≥3.5: sensitivity 84.5% [95% CI: 79.1%-90.3%] and specificity 83.2% [95% CI: 76.1%-90.3%] for TSB ≥11 mg/dL). Areas under the receiver operating characteristic curve for identifying TcB ≥11, ≥13, and ≥15 were 0.92, 0.93, and 0.94, respectively, and 0.90, 0.87, and 0.86 for identifying TSB ≥11, ≥13, and ≥15. Interrater reliability was high; 97% of scores by independent readers fell within 1 score of one another (N = 88). CONCLUSIONS: The Bili-ruler is a low-cost, noninvasive tool with high diagnostic accuracy for neonatal jaundice screening. This device may be used to improve referrals from community or peripheral health centers to higher-level facilities with capacity for bilirubin testing and/or phototherapy.