A recent study published in the journal Sensors introduces a novel wearable device designed to accurately and conveniently detect fetal movements. By integrating advanced sensors with IoT technology, this device provides continuous monitoring outside hospital settings, enhancing maternal care and enabling remote medical oversight.
Study: An IoT-Enabled Wearable Device for Fetal Movement Detection Using Accelerometer and Gyroscope Sensors. Image Credit: Barou abdennaser/Shutterstock.com
Download your PDF copy now!
Background
Ensuring effective prenatal care is critical, particularly in addressing declining birth rates and high fetal mortality rates seen in many countries. A key component of this care is monitoring fetal movements, which serves as an early indicator of potential health concerns and helps safeguard both mother and baby.
While various wearable systems exist for maternal care, many rely on complex processing methods or lack real-time feedback, limiting their practicality. The integration of IoT with wearable technology presents a compelling solution, enabling continuous data collection and transmission.
Although previous research has explored similar approaches, a dedicated, user-friendly system specifically designed for fetal movement detection remains necessary. This study introduces a device that addresses this need by combining real-time IoT capabilities with advanced machine learning algorithms, making fetal monitoring more efficient and accessible.
Device Overview
At the core of the device is a single Inertial Measurement Unit (IMU) that incorporates a three-axis accelerometer and a three-axis gyroscope. The accelerometer, capable of measuring accelerations up to ±16 g, and the gyroscope, with a range of ±2000°/second, work in tandem to capture precise fetal movement data.
The ESP-32 microcontroller powers the device, chosen for its efficient power management and integrated communication features, including Wi-Fi and Bluetooth Low Energy (BLE). Designed for ease of use, the lightweight device (approximately 25 g) adheres to the mother’s lower abdomen using medical-grade adhesive patches, ensuring comfort and reliability.
To refine classification accuracy, the researchers evaluated ten signal extraction methods, six machine learning algorithms, and four feature selection techniques.
A key method employed was Particle Swarm Optimization (PSO), which facilitated both feature selection and hyperparameter tuning. Among the machine learning techniques assessed, Extreme Gradient Boosting (XGB) emerged as the most effective classification method, achieving a sensitivity (SEN) of 90.00 %, precision (PRE) of 87.46 %, and an F1-score of 88.56 %.
Beyond local processing, the device transmits collected data to a cloud-based platform via the ESP-32 microcontroller, ensuring secure and efficient remote monitoring. The system upholds data integrity with an average communication latency of 423.6 milliseconds, allowing healthcare providers to access and review fetal movement data in real-time.
Results and Discussions
The device demonstrated strong accuracy in detecting fetal movements, as reflected in its high sensitivity and precision metrics. Its wireless functionality ensures timely data transmission to medical professionals, a critical factor in prenatal care. The integration of machine learning significantly improved differentiation between fetal and non-fetal movements, further enhancing monitoring reliability.
User feedback was overwhelmingly positive, with 35 pregnant participants reporting a high level of comfort while using the device. Extensive testing of various signal extraction methods and machine learning algorithms helped minimize false positives and negatives, reinforcing detection accuracy. Additionally, the PSO-based feature selection process optimized movement analysis, contributing to more precise monitoring outcomes.
By leveraging IoT technology, the device shifts prenatal monitoring from traditional clinical settings to a more modern, patient-centered approach. Remote access to fetal health data also allows healthcare providers to track fetal well-being without requiring hospital visits—an especially valuable advancement for individuals in remote or underserved areas.
Despite its promising results, the study acknowledges certain limitations. Broader data collection is needed to ensure the system’s effectiveness across diverse conditions. Additionally, further exploration of advanced classification techniques may enhance performance. Feedback from healthcare professionals could also guide refinements to the user interface of the accompanying mobile application, ultimately improving the experience for both patients and medical practitioners.
Conclusion
This study presents a promising wearable device for fetal movement detection, integrating accelerometer and gyroscope sensors with IoT technology. Its strong performance metrics highlight its potential for continuous and reliable operation, offering reassurance to both mothers and healthcare providers.
The findings validate the real-world application of this technology, marking a significant advancement in prenatal health monitoring. Future research will focus on refining the system further to address a broader range of maternal and fetal health needs.
Journal Reference
Rattanasak A., Jumphoo T., et al. (2025). An IoT-Enabled Wearable Device for Fetal Movement Detection Using Accelerometer and Gyroscope Sensors. Sensors 25(5):1552. DOI: 10.3390/s25051552, https://www.mdpi.com/1424-8220/25/5/1552