A Novel Human Embryo Microscope Image Classification Technique Based on ConvNeXtLarge Model

Abstract

Assisted reproductive technology (ART) has indeed revolutionized the field of infertility treatment and has significantly improved the chances of a successful pregnancy for many couples. In vitro fertilization (IVF) is one of the most well-known and widely used ART methods. This research highlights the importance of embryological analysis in the IVF process and the potential role of artificial intelligence, specifically deep learning algorithms, in improving this aspect of the procedure. ART methods help address infertility issues in couples through various medical procedures. IVF is one such technique where eggs and sperm are combined outside the body, allowing for fertilization and early embryo development in a controlled environment. The success of ART procedures, especially IVF, depends on the quality and morphology of the embryos. Embryos typically develop into blastocysts within 3-5 days. These blastocysts consist of several critical components, including trophectoderm, zona pellucida, blastocoel, and inner cell mass.