Machine Learning Methods in the Treatment Outcome Prediction

The purpose of this study is to develop an application that could help clinicians to improve the quality and efficiency of treatment. In order to get an appropriate result, we use neural networks that will be trained with a real data. This research contains a brief review based on the literature of the past 5-6 years - highlighting the most important articles - showing the idea behind neural networks and where they are used in medicine. The definition, structure and operation of neural networks are presented. Database, used in this study, was provided by a medical center dealing with reproductive problems. The main aim is to purpose the most effective treatment based on the age group of a woman and her hormones. Presented application could be developed further to increase the accuracy and to provide the opportunity to purpose treatment considering absolutely all factors that may have an impact on the onset of pregnancy. Key words: neural networks, treatment, accuracy, algorithms, prediction, application.