An IoT-based wearable system using accelerometers and machine learning for fetal movement monitoring

Fetal movement is an essential index of fetal well-being. However, existing technologies in clinical applications cannot provide an effective, long-term and easily accessible way for fetal movement monitoring. This paper presents a wearable system using accelerometers and machine learning for automatic monitoring of fetal movement. To realize the concept of the e-health home care, Internet of Things (IoT) is applied on the system to connect all the terminal monitoring units to a control center. The whole system mainly consists of two parts: the local monitoring unit and the remote health evaluation unit, respectively. The local monitoring unit includes: (1) a garment integrating accelerometers for data acquisition and an embedded system for signal processing and machine learning, (2) an Android-based local monitoring platform used for visualization of statistics on fetal health status based on information received from the garment via Bluetooth. The local monitoring unit is also designed to alarm the mother if fetal movements reduce significantly or if the intensity is weakened. The remote health evaluation unit consists of a cloud computing platform and an expert system. Connecting each terminal local monitoring unit via the Internet, the remote health evaluation unit allows an expert to access the information at distance for further medical supervision, make advanced diagnosis as well as take actions in case of emergency. The cloud computing platform is also designed for convergence of medical data from each individual patient to build a general medical database for scientific research. The proposed local fetal movement monitoring unit along with the remote health evaluation platform could be regarded as a good example of IoT application in the field of smart textile and home healthcare.

[1]  Boualem Boashash,et al.  Non-invasivemonitoring of fetal movements using time-frequency features of accelerometry , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[2]  Boualem Boashash,et al.  Passive detection of accelerometer-recorded fetal movements using a time-frequency signal processing approach , 2014, Digit. Signal Process..

[3]  Susan E. Callan,et al.  Maternal Perception of Foetal Movement Compared With Movement Detected By Real-Time Ultrasound: An Exploratory Study , 2010 .

[4]  T. Moore,et al.  A prospective evaluation of fetal movement screening to reduce the incidence of antepartum fetal death. , 1989, American journal of obstetrics and gynecology.

[5]  Jonathan Lai,et al.  Fetal movements as a predictor of health , 2016, Acta obstetricia et gynecologica Scandinavica.

[6]  Shigeko Horiuchi,et al.  Automated Software Analysis of Fetal Movement Recorded during a Pregnant Woman’s Sleep at Home , 2015, PloS one.

[7]  Stephen Grossberg,et al.  Fuzzy ARTMAP: A neural network architecture for incremental supervised learning of analog multidimensional maps , 1992, IEEE Trans. Neural Networks.

[8]  Abdulhamit Subasi,et al.  EEG signal classification using wavelet feature extraction and a mixture of expert model , 2007, Expert Syst. Appl..

[9]  Ravi Vaidyanathan,et al.  Performance of a wearable acoustic system for fetal movement discrimination , 2018, PloS one.

[10]  Michael J. Jenkins,et al.  Fetal Movement Measurement and Technology: A Narrative Review , 2017, IEEE Access.

[11]  Fuentes Miguel,et al.  Detecting fetal movements using non-invasive accelerometers: A preliminary analysis , 2010, 10th International Conference on Information Science, Signal Processing and their Applications (ISSPA 2010).

[12]  M. B. Malarvili,et al.  FETAL MOVEMENTS RECORDING SYSTEM USING ACCELEROMETER SENSOR , 2018 .

[13]  Julien Penders,et al.  Variable-length accelerometer features and electromyography to improve accuracy of fetal kicks detection during pregnancy using a single wearable device , 2017, 2017 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI).