Seminal Quality Prediction Using Deep Learning Based on Artificial Intelligence

Abstract

Fertility rates have dramatically decreased in the last two decades, especially in men. It has been described that environmental factors, as well as life habits, may affect semen quality. This paper evaluates the performance of different artificial intelligence (AI) techniques for classifying fertility dataset that includes the semen sample analysed according to WHO 2010 criteria and publicly available on UCI data repository. In this context, deep neural network (DNN) which involved in many studies in recent years is proposed to classify fertility dataset successfully. For the purpose of comparing the proposed method’s performance, Adaptive Neuro-Fuzzy Inference system (ANFIS) is also used for the classification problem. The results show that the performance of the DNN has the best with the average accuracy rate of 90.11%, and the results of the other ANFIS methods are also satisfactory.