Related Experiment Video
Updated: Jan 7, 2026
07:34
Fetal Mouse Cardiovascular Imaging Using a High-frequency Ultrasound 30/45MHZ System
Published on: May 5, 2018
12.1K
A machine learning model for assessing fetal health during pregnancy
Arshad Kalathil Ashik1, Robert Gutierrez1, Fidha Ashraf2
1Department of Mechanical Engineering, Imperial College London, London, United Kingdom.
Frontiers in Bioengineering and Biotechnology
|January 2, 2026
Summary
A new wearable sensor array effectively tracks fetal movements (FM), offering a potential low-cost solution to reduce stillbirths, especially in low- and middle-income countries (LMICs). This technology could improve perinatal outcomes by enabling timely interventions.
Area of Science:
- Biomedical Engineering
- Maternal-Fetal Medicine
- Signal Processing
Background:
- Stillbirth remains a critical global health issue, disproportionately affecting low- and middle-income countries (LMICs).
- Current methods for monitoring fetal movement (FM) lack objectivity and clinical impact, necessitating improved technologies.
- Existing wearable fetal movement monitors have shown limited clinical utility due to homogeneous sensing and lack of validation.
Purpose of the Study:
- To validate a novel multimodal vibrational-acoustic sensor array for accurate fetal movement (FM) monitoring.
- To assess the feasibility of a low-cost, wearable FM monitor for use in diverse healthcare settings, including LMICs.
Main Methods:
- A wearable sensor array with piezoelectric and acoustic modalities was developed and tested.
- 25 pregnant participants were enrolled, recording vibrational-acoustic data concurrently with ultrasound.
- Machine learning models, including RUSBoost, were employed to analyze sensor data and predict categorized fetal movements.
Main Results:
- The multimodal sensor array demonstrated feasibility in tracking fetal movements.
- An ensemble RUSBoost model achieved a precision of 0.44 and a recall of 0.61 for FM prediction.
- The study validates the potential of inexpensive, off-the-shelf sensors for a wearable FM monitor.
Conclusions:
- The validated vibrational-acoustic sensor array offers a promising foundation for developing effective wearable fetal movement monitors.
- This technology has the potential to significantly improve stillbirth detection and prevention, particularly in resource-limited settings like LMICs.
- Further development could lead to a widely accessible tool for enhancing perinatal care and reducing stillbirth rates globally.
