Detection of Male Fertility Using AI-Driven Tools

• Published in: Recent Trends in Image Processing and Pattern Recognition Vol. 1576; pp. 14 - 25
• Main Authors: Roy, Debasmita Ghosh, Alvi, P. A.
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2022; Springer International Publishing
• Series: Communications in Computer and Information Science
• Subjects: AI tools; Feature selection; Male fertility; Receiver Operating Characteristic (ROC)
• ISBN: 3031070046; 9783031070044
• ISSN: 1865-0929; 1865-0937
• DOI: 10.1007/978-3-031-07005-1_2

In the last few decades, the nation has been experiencing a low fertility rate due to fast changes in human lifestyle over a short period. Many lifestyle factors, such as liquor consumption, physical latency, cigarette smoking, caffeine intake, and others, can adversely affect on reproductive performance. These factors are associated with sperm quality, which is a pivotal key feature to identify male fertility status. In this experiment, three different feature selection methods have been applied to assess the uppermost features which are deeply connected with seminal quality. The final dataset contains three lifestyle features of hundred males under 18 to 36 years of age, having normal and altered output labels. Four artificial intelligence methods such as logistics regression, support vector machine, decision tree, and k-nearest neighbor are utilized to identify the male reproductive state. Finally, K-nearest neighbor algorithm has excelled in male fertility prognosis with 90% efficacy, and the receiver operating characteristic value is 0.85.