A Real-Time Oocyte Polar Body Detection Method

Abstract

Abstract Based on the extensive application requirements of cell micromanipulation in bioengineering, this paper focuses on the intracytoplasmic sperm injection technology of assisted reproductive technology in the field of obstetrics and gynecology, and develops algorithms to automatically and accurately identify the position of the polar body of the oocyte, laying a foundation for the realization of sperm and oocyte automated injection. This paper proposes an algorithm that can accurately locate polar body of oocyte; this method uses a lightweight U-Net network to semantically segment oocyte images and extract their edges, and then search for polar body based on image edges. This method has been verified on 524 pictures of mice oocyte, which is real-time and robust.