Abstract—Quality assessment of bull sperms is a necessary task to process to increase a fertilization rate when applying methods of Assisted Reproductive Technology including In Vitro Fertilization and Intracytoplasmic Sperm Injection. Presently, the development of Computer-Assisted Sperm Analysis (CASA) allows scientists to assess the quality of bull sperms automatically in some processes such as sperm motility. However, scientists commonly perform the viability assessment process manually that surely requires many endeavors and highly time-consuming. In addition, the standard commercial CASAs provide only a module for assessing sperm motility. Unfortunately, to add an extra module of the assessment of sperm viability to the standard CASA is much costly. In this study, we proposed a novel sperm detection and viability classification system that use to distinguish dead or alive cell of the bull sperms by using Faster Region-based Convolutional Neural Networks (Faster RCNN). The performance of the proposed system was evaluated in term of sperm detection and viability classification. The results of the performance evaluation process revealed that it provided a high accuracy for sperm detection (90.72%) as well as the viability classification (84.49%). Based on these results, it clarified that our proposed system had a great potential to use in the real-world scenarios.
 

Index Terms—sperm detection, viability classification, deep learning, faster RCNN

Cite: Sarunya Kanjanawattana, Prakaidoy Ditsayabut, Pumrapee Poomka, Kittipat Sriwong, Watthana Pongsena, and Chokchai Wanapu, "Viability Assessment of Bull Sperms Using Deep Learning," Journal of Advances in Information Technology, Vol. 12, No. 1, pp. 71-77, February 2021. doi: 10.12720/jait.12.1.71-77

Copyright © 2021 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.