Predictive Modeling of Recurrent Implantation Failure and Pre-eclampsia Using Machine Learning and Gene Expression Profiling

A pregnancy complication is any medical condition that arises during pregnancy that impacts the health of the mother, the fetus, or both. Recurrent implantation failure and pre-eclampsia are two such prenatal medical disorders. Machine learning systems can accurately predict high-risk prenatal medical conditions like recurrent implantation failure and pre-eclampsia. This study aimed to analyze differentially expressed genes for both pregnancy complications and develop a Machine learning model for the early prognosis of recurrent implantation failure and pre-eclampsia. Differentially expressed genes for recurrent implantation failure consisted of 2486 downregulated genes and 809 upregulated genes, and pre-eclampsia, consisted of 13 downregulated genes and 10 upregulated genes followed by gene set enrichment analysis. Gene expression prolife of recurrent implantation failure and pre-eclampsia were used to develop the machine learning model. Random Forest performed best for recurrent implantation failure with a model accuracy of 96.47%, while the generalized linear model performed best for pre-eclampsia with a model accuracy of 80%.