Smart Healthcare: Early Diagnosis of Fetal Health Conditions using Machine Learning and Cardiotocographic Data

Fetal health and early diagnosis refer to fetal development and regular contact with the mother’s uterus during pregnancy. Maximum pregnancy period complications put the fetus in a challenging situation that restricts proper growth and can result in deficiencies or death. Predicting risk factors before problems arise can help ensure a healthy pregnancy and promote appropriate fetal development. Forecasting the fetal health and growth state using a set of previously recognized patterns when creating a predictive classifier model with machine learning algorithms is essential. A fetus is an embryonic form of an unborn child that has not yet entered the world. Every three months of pregnancy are referred to as trimesters. The fetus grows and develops during this process, and regular examinations are crucial to ensure its health. There are many potential factors that could result in the newborn’s impairment or even death, which is a very serious situation that needs to be avoided. A Cardiotocography (C.T.G.), which is typically used to assess the heartbeat and uterine contractions, is one of the main techniques to assess the health of the fetus inside the womb. The doctor uses the data produced by C.T.G. test to assess the fetus state of his health and his words. However, there is space for error, making the doctors' data analysis untrustworthy. To assess the data and forecast fetal health, various machine learning and deep learning algorithms have been used. In this paper to demonstrate prediction accuracy using several categorization models and to evaluate the performance of each model.