IVF data analysis with data mining techniques

In this paper we develop data mining techniques incorporating the association rules and classification algorithms to search for the outcomes of IVF treatments when multiple fetal heart beats observed on ultrasound either as a result of the transfer of two embryos or homozygotes. Our objective is to detrmine the fate of the surviving fetus after the demise of twin. The data is large and and includes both the numerical and categorical or symbolic data. We extend our algorithm to incorporate clinical significance by way of introducing a quasi fuzzy network. Our results show that these techniques produce more clear picture than the ones achieved through traditional statistical methods.