*Result*: Machine Learning Techniques to Predict Fertility Rate of Sperm from the Outcome of IVF Functional Tests.

*The objectives of this study are: (i) to predict the fertilization potential of animal sperm from the outcome of In Vitro sperm Function (IVF) tests using a machine learning technique of Artificial Neural Network (ANN), combined with the backpropagation algorithm; and (ii) to compare the predictive accuracy of the resulting model with that of a statistical Multi-Linear Regression (MLR) model. Artificial Insemination (AI) is one of the most successful reproductive technologies developed to improve the reproductive efficiency of farm animals and dairy cattle. The percentage of pre-freeze motility and the assays of sperm functions such as acrosome reaction, zona binding ability, in vitro fertilization, and in vitro embryo production are used to predict fertility in the field. ANN offers a novel approach to pattern recognition and can be more effective than traditional statistical techniques for identity associations, due, in part, to their ability to recognize highly non-linear associations (Durairaj and Meena, 2008). The work was carried out by using the MATLAB and Nuero Solutions software. The fertility rate of sperm was predicted using the feedforward multilayer perceptron ANN. Backpropagation algorithm was used for training the ANN in MATLAB. The results show that the predictive accuracy of the devised ANN (r = 0.995) is higher than that of the traditional Non-Linear Regression (NLR) (r = 0.903) model. [ABSTRACT FROM AUTHOR]*