A mobile fetal health monitoring device

Abstract

Maternal and infant mortality is unacceptably high in many parts of the developing world. About 830 women die from pregnancy- or childbirth-related complications around the world every single day. The world health organization (WHO) estimated that in 2018, roughly 303 000 women died during and following pregnancy and childbirth. Almost all of those deaths occurred in low-resource settings, yet most of these fatal cases can be prevented. One of the key causes of this death is limited access to quality antenatal care in part due to; a limited number of medical professionals, lack of appropriate tools and data infrastructure to monitor the progress of the pregnancies, and detected complicated cases early enough.

Therefore, this research aimed at developing a home-based mobile fetus health monitoring device to help expectant mothers to monitor the health status of the fetus as part of an effort to improve early detection of pregnancy complications. The mobile device is based on robust multifactor prediction models which use physiological sensors for the signal sampling of various parameters. The device is designed to be simple to use for even semi-literate mothers who use common user interface design principles. The device is designed to be low cost to enable wide accessibility by the target community. The design and evaluation of this developed mobile fetus health monitor device followed a design science research approach to ensure that it addresses the needs of the stakeholders in the problem domain (that is the expectant mothers and Midwives). The devices requirements were established through; medical professional’s expert interviews, target end-user exploratory interviews and observations, literatures reviews and stakeholder design thinking workshops. The hardware is designed using non-invasive sensors on a wearable belt that uses phonocardiogram and electromyogram principles in capturing physiological signals from mother and fetus following embedded system principles and the software is designed using open sources technologies following embeds system.

The device was evaluated for system correctness through function tests, usefulness through experimentations, and usability through expert and field study. The results of function testing revealed that the algorithms implementation was done well and produces the expected results. Furthermore, the result of correctness revealed that the device was able to provide precision accuracy of 88% for fetal heart rate, 95% for contraction strength 93% for contraction duration, and 92% for contraction frequency. In terms of usefulness, the results of the field trial revealed that the device was able to detect complications early and trigger action in seeking care for expectant mothers. The mothers appreciated the simplicity of the interface and 77% indicated that results are easy to interpret. In terms of usability, 67% of the end-users indicate the device to be easy to learn, 80% indicated to be easy to use, and overall, 90% of experts indicated the device was easy to learn and use. The results of the field trials show that the device was effective in detecting accurately 93% of the health status of the fetus in comparison with the human expert. However, like any medical device, more work is needed to evaluate in wider populations, but also to continuously improve the sampling algorithm to obtain a lower signal to noise rate.