Embryo Selection for IVF using Machine Learning Techniques Based on Light Microscopic Images of Embryo and Additional Factors.

The current process of embryo selection in In Vitro Fertilization (IVF) process is based on morphological criteria, e.g., Istanbul scoring system and manually evaluated by embryologists; consequently, the assessment can be subjective. In the case of multiple embryos that have the same morphological grading, there is no guidance on how embryos should be prioritized to be transferred. This work aims to develop a deep learning-based model to classify viable and non-viable embryos using light microscopic images of an embryo. Additional features according to Istanbul grading system and the patients' age is also included in the model. Various models are evaluated and the best model based on the fusion of embryo images and additional features provides accuracy, sensitivity, and area under curve (AUC) of 65%, 74.29% and 0.72, respectively. The distributions of the prediction score corresponding to each additional feature are analysed and compared with pregnant and non-pregnant ground truths. We have found that the additional factors, such as age, embryo development stage, the quality of inner cell mass (ICM), and trophectoderm (TE) have a positive impact and enhanced the model prediction of implantation potential.

Human follicular fluid elicits select dose- and age-dependent effects on mouse oocytes and cumulus-oocyte complexes in a heterologous in vitro maturation assay.

Follicular fluid (FF) is a primary microenvironment of the oocyte within an antral follicle. Although several studies have defined the composition of human FF in normal physiology and determined how it is altered in disease states, the direct impacts of human FF on the oocyte are not well understood. The difficulty of obtaining suitable numbers of human oocytes for research makes addressing such a question challenging. Therefore, we used a heterologous model in which we cultured mouse oocytes in human FF. To determine whether FF has dose-dependent effects on gamete quality, we performed in vitro maturation of denuded oocytes from reproductively young mice (6-12 weeks) in 10%, 50%, or 100% FF from participants of mid-reproductive age (32-36 years). FF impacted meiotic competence in a dose-dependent manner, with concentrations >10% inhibiting meiotic progression and resulting in spindle and chromosome alignment defects. We previously demonstrated that human FF acquires a fibro-inflammatory cytokine signature with age. Thus, to determine whether exposure to an aging FF microenvironment contributes to the age-dependent decrease in gamete quality, we matured denuded oocytes and cumulus-oocyte complexes (COCs) in FF from reproductively young (28-30 years) and old (40-42 years) participants. FF decreased meiotic progression of COCs, but not oocytes, from reproductively young and old (9-12 months) mice in an age-dependent manner. Moreover, FF had modest age-dependent impacts on mitochondrial aggregation in denuded oocytes and cumulus layer expansion dynamics in COCs, which may influence fertilization or early embryo development. Overall, these findings demonstrate that acute human FF exposure can impact select markers of mouse oocyte quality in both dose- and age-dependent manners.

Cumulus expansion is impaired with advanced reproductive age due to loss of matrix integrity and reduced hyaluronan.

Reproductive aging is associated with ovulatory defects. Age-related ovarian fibrosis partially contributes to this phenotype as short-term treatment with anti-fibrotic compounds improves ovulation in reproductively old mice. However, age-dependent changes that are intrinsic to the follicle may also be relevant. In this study, we used a mouse model to demonstrate that reproductive aging is associated with impaired cumulus expansion which is accompanied by altered morphokinetic behavior of cumulus cells as assessed by time-lapse microscopy. The extracellular matrix integrity of expanded cumulus-oocyte complexes is compromised with advanced age as evidenced by increased penetration of fluorescent nanoparticles in a particle exclusion assay and larger open spaces on scanning electron microscopy. Reduced hyaluronan (HA) levels, decreased expression of genes encoding HA-associated proteins (e.g., Ptx3 and Tnfaip6), and increased expression of inflammatory genes and matrix metalloproteinases underlie this loss of matrix integrity. Importantly, HA levels are decreased with age in follicular fluid of women, indicative of conserved reproductive aging mechanisms. These findings provide novel mechanistic insights into how defects in cumulus expansion contribute to age-related infertility and may serve as a target to extend reproductive longevity.

Morphokinetic parameters of mouse oocyte meiotic maturation and cumulus expansion are not affected by reproductive age or ploidy status.

INTRODUCTION: Morphokinetic analysis using a closed time-lapse monitoring system (EmbryoScope + ™) provides quantitative metrics of meiotic progression and cumulus expansion. The goal of this study was to use a physiologic aging mouse model, in which egg aneuploidy levels increase, to determine whether there are age-dependent differences in morphokinetic parameters of oocyte maturation. METHODS: Denuded oocytes and intact cumulus-oocyte complexes (COCs) were isolated from reproductively young and old mice and in vitro matured in the EmbryoScope + ™. Morphokinetic parameters of meiotic progression and cumulus expansion were evaluated, compared between reproductively young and old mice, and correlated with egg ploidy status. RESULTS: Oocytes from reproductively old mice were smaller than young counterparts in terms of GV area (446.42 ± 4.15 vs. 416.79 ± 5.24 µm2, p < 0.0001) and oocyte area (4195.71 ± 33.10 vs. 4081.62 ± 41.04 µm2, p < 0.05). In addition, the aneuploidy incidence was higher in eggs with advanced reproductive age (24-27% vs. 8-9%, p < 0.05). There were no differences in the morphokinetic parameters of oocyte maturation between oocytes from reproductively young and old mice with respect to time to germinal vesicle breakdown (GVBD) (1.03 ± 0.03 vs. 1.01 ± 0.04 h), polar body extrusion (PBE) (8.56 ± 0.11 vs. 8.52 ± 0.15 h), duration of meiosis I (7.58 ± 0.10 vs. 7.48 ± 0.11 h), and kinetics of cumulus expansion (0.093 ± 0.002 vs. 0.089 ± 0.003 µm/min). All morphokinetic parameters of oocyte maturation were similar between euploid and aneuploid eggs irrespective of age. CONCLUSION: There is no association between age or ploidy and the morphokinetics of mouse oocyte in vitro maturation (IVM). Future studies are needed to evaluate whether there is an association between morphokinetic dynamics of mouse IVM and embryo developmental competence.

Quantitative morphokinetic parameters identify novel dynamics of oocyte meiotic maturation and cumulus expansion†.

Meiotic maturation and cumulus expansion are essential for the generation of a developmentally competent gamete, and both processes can be recapitulated in vitro. We used a closed time-lapse incubator (EmbryoScope+™) to establish morphokinetic parameters of meiotic progression and cumulus expansion in mice and correlated these outcomes with egg ploidy. The average time to germinal vesicle breakdown (GVBD), time to first polar body extrusion (PBE), and duration of meiosis I were 0.91 ± 0.01, 8.82 ± 0.06, and 7.93 ± 0.06 h, respectively. The overall rate of cumulus layer expansion was 0.091 ± 0.002 µm/min, and the velocity of expansion peaked during the first 8 h of in vitro maturation (IVM) and then slowed. IVM of oocytes exposed to Nocodazole, a microtubule disrupting agent, and cumulus oocyte complexes (COCs) to 4-methylumbelliferone, a hyaluronan synthesis inhibitor, resulted in a dose-dependent perturbation of morphokinetics, thereby validating the system. The incidence of euploidy following IVM was >90% for both denuded oocytes and intact COCs. No differences were observed between euploid and aneuploid eggs with respect to time to GVBD (0.90 ± 0.22 vs. 0.97 ± 0.19 h), time to PBE (8.89 ± 0.98 vs. 9.10 ± 1.42 h), duration of meiosis I (8.01 ± 0.91 vs. 8.13 ± 1.38 h), and overall rate and kinetics of cumulus expansion (0.089 ± 0.02 vs 0.088 ± 0.03 µm/min) (P > 0.05). These morphokinetic parameters provide novel quantitative and non-invasive metrics for the evaluation of meiotic maturation and cumulus expansion and will enable screening compounds that modulate these processes.

Influence of hydrogel encapsulation during cryopreservation of ovarian tissues and impact of post-thawing in vitro culture systems in a research animal model.

OBJECTIVE: Using domestic cats as a biomedical research model for fertility preservation, the present study aimed to characterize the influences of ovarian tissue encapsulation in biodegradable hydrogel matrix (fibrinogen/thrombin) on resilience to cryopreservation, and static versus non-static culture systems following ovarian tissue encapsulation and cryopreservation on follicle quality. METHODS: In experiment I, ovarian tissues (n=21 animals; 567 ovarian fragments) were assigned to controls or hydrogel encapsulation with 5 or 10 mg/mL fibrinogen (5 or 10 FG). Following cryopreservation (slow freezing or vitrification), follicle viability, morphology, density, and key protein phosphorylation were assessed. In experiment II (based on the findings from experiment I), ovarian tissues (n=10 animals; 270 ovarian fragments) were encapsulated with 10 FG, cryopreserved, and in vitro cultured under static or non-static systems for 7 days followed by similar follicle quality assessments. RESULTS: In experiment I, the combination of 10 FG encapsulation/slow freezing led to greater post-thawed follicle quality than in the control group, as shown by follicle viability (66.9%±2.2% vs. 61.5%±3.1%), normal follicle morphology (62.2%±2.1% vs. 55.2%±3.5%), and the relative band intensity of vascular endothelial growth factor protein phosphorylation (0.58±0.06 vs. 0.42±0.09). Experiment II demonstrated that hydrogel encapsulation promoted follicle survival and maintenance of follicle development regardless of the culture system when compared to fresh controls. CONCLUSION: These results provide a better understanding of the role of hydrogel encapsulation and culture systems in ovarian tissue cryopreservation and follicle quality outcomes using an animal model, paving the way for optimized approaches to human fertility preservation.