Experimental Evaluation of Multi-Quadrilateral-Pillar Array Microfluidic Platform for Oocyte Denudation Utilizing a Deep-Learning Approach

Oocyte denudation is the procedure of removing the cumulus cells encasing an oocyte to expose the oocyte itself. During oocyte maturation, cumulus cells play a crucial role in providing nutritional factors. However, it is necessary to isolate cumulus cells from the cumulus-oocyte complex (COC) to assess the oocyte maturity and enable procedures like intracytoplasmic sperm injection (ICSI). Traditional oocyte denudation procedures are often labor intensive and operator dependent, leading to variability in denudation efficacy and potential mechanical damage to the oocyte during manual pipetting. In this work, we designed and experimented with a novel microfluidic system for oocyte denudation using a multi-quadrilateral-pillar array structure. The micropillar array was configured to generate sharp shear angles that mechanically disrupted the connections between the oocyte and surrounding cumulus cells as the COC passed through while also creating a zigzag flow trajectory that facilitates thorough denudation across different regions of the oocyte surface. Image processing and computer vision techniques were applied to evaluate denudation efficiency, confirming that the chip performed effectively in removing cumulus cells from porcine COCs. The proposed denudation on-chip shows strong potential for integration into automated and streamlined workflows in assisted reproductive technologies (ART). Its modular micropillar array can be expanded to increase COC throughput, enabling rapid and efficient large-scale oocyte denudation.