Male Infertility Prediction Model Using Aritificial Neural Network in Surabaya, Indonesia

Abstract

In this study, a fertility clinic in Surabaya, Indonesia, uses artificial neural networks (ANNs) to detect male infertility problems. Despite improvements in fertility diagnostics, there are still issues with precisely forecasting infertility from a variety of patient data collected by non-specific entities. In addition to being inconvenient, male fertility diagnostic techniques like semen analysis, sperm function testing, hormone testing, and genetic testing can also cause discomfort and emotional distress for many patients. The research utilizes a dataset of 260 male patients, divided into training (208 samples) and testing (52 samples) sets, to develop predictive models. Employing a backpropagation neural network (BPNN) model, the study achieved a prediction accuracy performance of 96.6%, highlighting the model's effectiveness in identifying abnormalities in semen parameters linked to male infertility. Key parameters influencing predictions included sperm concentration and morphology, with hypospermia emerging as a significant factor. The results demonstrate that BPNNs can enhance diagnostic precision and facilitate tailored treatment plans for patients, addressing the limitations of traditional diagnostic methods. This innovative approach not only contributes to the understanding of male infertility but also emphasizes the importance of integrating advanced technologies in reproductive health diagnostics. The findings suggest that the implementation of predictive models like BPNNs can significantly improve clinical outcomes for couples facing infertility challenges, paving the way for further research and application in this critical area of healthcare.