NurtureNest: an IoT-wearable predictive analytics framework for real-time maternal risk assessment

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: Test accuracy: 92.86%; Cross-validation accuracy: 95.04%

• Primary Outcome: Multi-class classification of pregnancy risk into low, medium, and high categories using LightGBM ensemble classifier

• Nnt Or Sensitivity: Sensitivity, specificity, and likelihood ratios not reported. Per-class diagnostic performance metrics absent. Overall accuracy figures only — clinically insufficient for diagnostic test evaluation

• Confidence Interval: Not reported

Clinical Application

Feasibility in clinical settings is undemonstrated. The framework requires smartwatch wearables and a mobile application, raising questions about device compatibility, data security, regulatory approval, patient digital literacy, connectivity infrastructure, and integration with existing electronic medical record systems. No real-world deployment or pilot data are presented This framework has no demonstrated applicability to Australian clinical practice at this stage. It has not been validated against Australian obstetric populations or risk factor prevalence. The TGA would require prospective clinical validation data before any such tool could be approved as a Software as a Medical Device (SaMD) under the TGA's Digital Health regulatory framework. RACGP and RANZCOG guidelines for antenatal risk assessment rely on validated clinical tools (e.g., validated pre-eclampsia risk calculators, standardised antenatal care pathways) that this framework has not been benchmarked against. PBS considerations are not relevant at this developmental stage. Aboriginal and Torres Strait Islander maternal health contexts, where risk factor profiles differ significantly, are not addressed. The framework may have greater near-term relevance in low-resource international settings if prospectively validated, but Australian clinical adoption would require substantial additional evidence Cannot be determined. The study population is not described. Intended use in pregnant women requiring risk stratification, but no demographic, gestational age, comorbidity, or prevalence data are provided to define the applicable population

Abstract

Pregnancy-related complications are increasing globally, necessitating timely and accurate risk prediction for effective clinical intervention. This paper presents NurtureNest, an internet of things-enabled machine learning framework for automated pregnancy risk assessment. The system integrates wearable sensor data from smartwatches with user-provided clinical parameters via a mobile application, enabling continuous remote monitoring. Historical clinical data are used to train multiple machine learning models for multi-class classification of pregnancy risk into low-, medium-, and high-risk categories. Ensemble learning techniques are independently trained and evaluated alongside conventional machine learning classifiers to assess their effectiveness in pregnancy risk prediction. Experimental results show that LightGBM achieves the highest performance with 92.86% test accuracy and 95.04% cross-validation accuracy. Model performance is validated using ROC analysis and feature importance evaluation. The proposed framework enables early risk detection and supports timely clinical decision-making, improving maternal healthcare outcomes.

References

1. 1.Rani, D., Wala, T., Kumar, R., & Chauhan, N. (2026). NurtureNest: an IoT-wearable predictive analytics framework for real-time maternal risk assessment. Biomedical Physics & Engineering Express. https://doi.org/10.1088/2057-1976/ae8305