An Investigation into Techniques used for Fetal Health Classification

Abstract

The natural birth of a mentally and physically sound child is the yearning of all mothers. Still, perinatal mortality is a huge concern that needs immediate heed. Prenatal attention towards the well-being of the mother and the child plays a vital role in this regard. Early detection of any abnormalities can give further insights into the pregnancy and will provide more time to parents and doctors to prepare for these unnatural circumstances. Cardiotocography (CTG) is a technique used for monitoring fetal heart rate, it is widely used to ensure fetal well-being during pregnancies at high risk. Usage of machine-learning techniques can automate this task and can reduce the chances of diagnostic errors. Deep Learning also has powerful algorithms for learning complicated characteristics and higher-level semantics. The principal objective of this paper is to dissect the boundaries of different classification algorithms and contrast their prescient exactnesses to discover the best classifier for ordering fetal wellbeing