An intelligent saliva recognition system for women’s ovulation detection

Abstract

This study presents using image processing and data mining approaches to develop a saliva image automatic recognition system for woman's ovulation prediction. Detect woman's ovulation can sterility treatment, diagnose certain diseases and avoid undesired pregnancies. Saliva ferning test is a technique that monitors woman's saliva and looks for patterns related to ovulation. We use a digital camera with a 100-time microscope to take saliva images. In the proposed method, six important features in dried saliva images are automatically extracted by employing some image processing techniques at first, and an intelligent system is developed by using the decision tree J48 algorithm for the detection of ovulation. In this study, the result of the best classification accuracy is 84% in 100 saliva samples. The proposed method has very important aspects of human, medical and economical. In addition, the proposed system can detect woman’s ovulation more safe, natural, convenient and efficient.