Robotic Piezo-Assisted Oocyte Penetration and Intracytoplasmic Sperm Injection

Abstract

Cell penetration and intracellular injection are indispensable procedures in many cell surgery tasks. Because of the complex layered structure of oocytes, conventional piezo-assisted penetration often causes unavoidable cellular damage. Moreover, the small volume of single cells and the nonlinear dynamics involved in the injection process make it challenging to achieve precise and rapid delivery of the injected material within the cytoplasm. This letter presents an integrated robotic system for automated oocyte penetration and intracellular sperm injection that enhances precision and minimizes cellular damage. A deep neural network enables robust segmentation of oocyte structural layers under low-contrast and partially occluded conditions, providing reliable feedback for penetration optimization. A dynamic model was developed to describe the interaction between the pump motion and the fluidic response. A model-predictive controller is then designed to compensate for delays and pressure deviations, ensuring smooth and accurate sperm delivery. Experimental validation demonstrates that the proposed system achieves penetration with a deformation of 4.76 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\pm$</tex-math></inline-formula> 2.34 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\mu$</tex-math></inline-formula>m and precise sperm positioning within <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\pm$</tex-math></inline-formula>5 pixels (overshoot below 8 pixels), while ensuring a post-injection survival rate of 82.8%.