Artificial Intelligence Techniques for the Detections of Congenital Diseases: Challenges and Research Perspectives

Abstract

Congenital infections, disorders, or diseases occur when pregnant women get infected with an organism that further enters into their placenta and fetus after entering their bloodstream. Such conditions may affect newborn infants and unborn fetuses that need to be cured by early prediction. In the field of medicine, better high-performance prediction is achieved using Artificial intelligence. It is a broad area of science for simulating the natural intelligence established by animals and humans through machine learning. In this paper, we have given an overview of using AI algorithms to predict various congenital diseases and forward with a broad category of congenital disorders and infections that come under prenatal and neonatal categories. Later, a comparative table was formulated to study its effects on the embryo and fetus. In the case of performing tasks using ML or DL algorithms, specific steps have been followed that are given in detail under the framework section. In the last section of the paper, we have performed a comparative study on work done in predicting different congenital diseases using fuzzy logic, deep neural network, Ensemble learning, Support vector machine, Artificial neural network, Random forest, and Naïve Bayes. From the study, it has been found that although different ML or DL algorithms used in the individual prediction of different congenital diseases can give good outcomes, more work still needs to be done on proposing approaches for predicting hybrid methods.