Ranking of Female Factors in Implantation Using Data Mining Techniques

Backgournd: In spite of improvements in infertility treatment, pregnancy rates have not increased significantly. Assisted reproductive technologies (ART) include costly and complexity processes. The aim of this study is determine the attributes and their particular values affecting the outcome in ART. Method: In this cross-sectional study, the data of 367 patients were collected using census method. The dataset contains 24 variables along with an identifier for each patient that is either negative or positive. To determine the significance of the female features, ranking-based algorithms such as Gain ratio and Gini Index run in Orange data mining software. Results: The results revealed the endometriosis is the importance factor among female pathologies. Our findings also demonstrate that the Infertility duration has highest score in Implantation outcomes. Conclusion: Elicited decision rules of ranking algorithms determine useful predictive features of Implantation. Out of 24 factors, the Infertility duration Thrombophilic disorders, and the status of period (Means) are the three best features for such prediction.

Keywords:

Clinical Decision Support , Data Mining , ranking algorithms , Assisted reproductive technologies (ART) , Predictive factors