Estimation of active sperm count in spermiogram using motion detection methods

Abstract

Nowadays, infertility is a serious health problem. The most important parameters that cause infertility in males are sperm morphology, sperm motility and sperm density. For this reason, sperm motility, density and morphology are analyzed in semen analysis and these analyzes are done by experts in the laboratory. Analysis based on observation in the laboratory is easy to mistake, subjective application. In this study, computer aided sperm count estimation method is suggested to minimize the effect of experts in semen analysis. In this study, especially the sperm motility is focused and the number of active sperm in the semen is estimated by using motion detection methods. Tests of the proposed method were performed with the data set prepared by the Isfahan Fertility and Infertility Center (IFIC). The experimental results obtained are compared with the previously proposed methods in the literature and it is seen that the proposed method gives successful results.

Publication
Journal of the Faculty of Engineering and Architecture of Gazi University, 2018(18–2)
Abdulkadir Gumuscu
Abdulkadir Gumuscu
Associate Professor of Artificial Intelligence

My research focuses on machine learning, computer vision and optimization for health, agriculture and energy.