Selection of competent oocytes by morphological criteria for assisted reproductive technologies

Abstract

Invasive and noninvasive methods are commonly used to select developmentally competent oocytes that can improve the take-home baby rates in assisted reproductive technology (ART) centers. One of the noninvasive methods conventionally utilized to determine competent oocytes is the morphological analysis of cumulus complex, first polar body, zona pellucida, perivitelline space, meiotic spindle, and ooplasm. Successful fertilization, early embryo development, uterine implantation, and healthy pregnancy depend on the quality of oocytes, and morphological evaluation is one of the options used to predict quality levels. In this review, the morphological criteria being utilized in certain ART centers are comprehensively evaluated with special references to their predictive values and potential contributions to selecting high-quality oocytes.

An interpretable and versatile machine learning approach for oocyte phenotyping

“…These features described the oocyte, its surrounding zona pellucida, its perivitelline space (space between the oocyte and its zona pellucida) and its dynamics (Figure 2C). For this, we used radiomics features (Fornacon-Wood et al, 2020; Rizzo et al, 2018), morphological characteristics that have been shown to be important to discriminate oocyte quality (Inoue et al, 2007; Ozturk, 2020; Rienzi et al, 2011), and other characteristics that could be applied to oocyte description. We implemented the measures of all features based on the segmentation of oocyte and zona pellucida contours from images acquired in transmitted light using our plugin (Supplemental file 1).…”

“…Moreover, this feature was also relevant for predicting human oocyte maturation potential. This texture could be related to the birefringence of the zona pellucida, which is used to assess human embryo quality (Gabrielsen et al, 2000; Ozturk, 2020; Raju et al, 2007; Rienzi et al, 2011) and the reflect the structuration of the zona. Cytoplasmic particles were smaller in oocytes with a low competence but also in Fmn2 -/- oocytes (Figure 3D), so this difference observed in the control oocyte population could be related to differences in the cytoplasmic actin activity that was reduced both in oocytes of low competence score and in Fmn2 -/- oocytes (Figure S5B, Figure 3C).…”

“…We decided to favor interpretability over predictive power and chose a machine learning approach based on interpretable features extracted from our plugin. We designed a list of 118 morphological features identified as potential markers of oocyte quality (Ozturk, 2020; Rienzi et al, 2011). We showed that this machine learning approach can be used to discriminate oocyte populations and automatically identify morphological differences coming from a mutant strain.…”

Digital PCR quantification of DNA, RNA and extracellular microRNA of mouse oocytes

“…The ability to conduct accurate and sensitive oocyte genetic analysis is critical for gaining insights to oocyte maturation, development, aging, and oocyte production in vitro, serving as a basis for understanding infertility disorders, determining fertility outcomes [58]. In addition, our findings underpin the importance of oocyte selection for assisted reproductive techniques or ART, whereby multiple oocytes are harvested from which one oocyte would be selected for artificial insemination, paving the way for increasing the chance for implantation [59].…”

Methods for Assessing Oocyte Quality: A Review of Literature

“…Thus, the place of assisted reproductive technology (ART) grows every day. However, despite the development of ever more efficient and precise techniques, the success rate in ART remains at about 35% in terms of pregnancy per embryo transferred [ 1 ]. Since the quality of the oocyte determines the embryonic development potential, this unique and highly specialized cell takes an important place in research to improve ARTs results.…”