Information and communication technology today enables health organizations to contact marginalized people in remote areas using sensing and artificial intelligence technologies. Applications of these technologies are even more crucial for maternal and newborn health because they are essential for a healthy society. Over the past few years, scientists have been researching sensing and artificially intelligent healthcare systems for mother and baby health. Sensors are utilized to monitor patient health parameters. The wearable sensors and AI algorithms mentioned in this chapter are based on existing systems and are designed to estimate risk factors for both mothers and children before, during, and after pregnancy. The sensors and AI algorithms employed in these systems are included in this review, which also examines each approach's characteristics, results, and innovative aspects in chronological sequence. Additionally, it discusses the datasets that were used, additional difficulties, and possible future paths for research.
Top1. Introduction
A healthy society depends on maternal and newborn health. Extra medullary pregnancy, premature birth, preeclampsia-causing hypertension, and labor inability that could result in a cesarean section are the main complications that affect mother health and even result in mortality. A major contributor to prenatal and postnatal health issues that might result from antepartum or postpartum hemorrhage, retained placenta, vaginal infection following birth, and many other comparable issues is iron deficiency abnormalities. A newborn's health may be significantly impacted by a number of major conditions, including uterine growth retardation, birth asphyxia, and infection from the mother’s vaginal delivery at the time of delivery, shoulder dystocia, meconium aspiration, septicemia, premature delivery /premature lungs, and any other congenital abnormalities. Maternal and newborn health issues must be recognized and informed about as soon as possible so that specialists can take appropriate action. However, it can be challenging to reach medical help in time, particularly in remote locations. Information and communication technology has made it possible for the medical community to quickly address issues with mother and infant health in the modern era.
Technology advancements enable the pharmaceutical profession to provide improvisatory prenatal and pediatric welfare services employing sensors, artificial intelligence, and computing platforms right at the patient's door. As the Internet of Things expands, Computational models and sensor-enabled systems can both recognize genetic materials use computational platforms offer enhanced communication, massive storage, and computing capabilities for the data that will be generated at an exponential rate. This information can be used to anticipate a person's health (IoTs) (Hamil et al., 2022).
Machine learning (ML)-based methods may be able to anticipate aberrant behavior in terms of mother and baby health. The huge growth of ML algorithms to track mother and baby health throughout the early stages of pregnancy may help doctors address issues. ML methods are currently being used to treat postpartum depression, diagnose congenital heart disease in pregnant women, detect wild stress, identify prenatal danger, and forecast the risk of preterm delivery. Additionally, the ML approaches have the capacity to track the infant's brain development, general growth, and health status. A variety of techniques, including Support Vector Machine (SVM), Artificial Neural Networks (ANN), Regression Analysis (RA), and Random Forest, are being utilized to determine the most effective maternal health (Al-Hassani & Atilla, 2023). As a result, it has been deemed to be the cornerstone of the dissemination of maternity and baby care as interest in the potential of ML in medical services, particularly for maternal and child health, has recently increased.
Wearable sensor-based technologies and ML approaches have the potential to successfully replace traditional healthcare settings by improving patient-provider interactions for effective pregnant health management. In order to improve pregnancy problem detection and monitor baby health, this article looks at existing wearable sensing technologies and machine learning techniques. The contributions of our research to this survey can be summed up as follows: Each healthcare framework, system, or model is examined in reverse chronological order. Discussions about the properties, size, format, and other details of real-world datasets are also included. A thorough review of the issues and opportunities in current research relating to these systems is also included, as is an examination of potential career paths for researchers in the relevant field.
This has been organized the following manner, section 1 discuss in detail about the detailed introduction about the wearable biosensors and AI algorithm. In section 2, detailed case study of the literature was done. Section 3, analysis and discussion of wearable bio-sensors. Section 4, discuss in detail about the impact of AI algorithms towards health monitoring of pregnant women’s. Section 5, proposes architecture for smart pregnancy management. Section 6, discuss about the conclusion and future work of the proposed architecture.