Approaches to reliably predict the developmental potential of embryos and select suitable embryos for blastocyst culture are needed. The development of time-lapse monitoring (TLM) and artificial intelligence (AI) may help solve this problem. Here, we report deep learning models that can accurately predict blastocyst formation and usable blastocysts using TLM videos of the embryo’s first three days. The DenseNet201 network, focal loss, long short-term memory (LSTM) network and gradient boosting classifier were mainly employed, and video preparation algorithms, spatial stream and temporal stream models were developed into ensemble prediction models called STEM and STEM⁺. STEM exhibited 78.2% accuracy and 0.82 AUC in predicting blastocyst formation, and STEM⁺ achieved 71.9% accuracy and 0.79 AUC in predicting usable blastocysts. We believe the models are beneficial for blastocyst formation prediction and embryo selection in clinical practice, and our modeling methods will provide valuable information for analyzing medical videos with continuous appearance variation.

Progressive loss of physiological integrity and accumulation of degenerative changes leading to functional impairment and increased susceptibility to diseases are the main features of aging. The ovary, the key organ that maintains female reproductive and endocrine function, enters aging earlier and faster than other organs and has attracted extensive attention from society. Ovarian aging is mainly characterized by the progressive decline in the number and quality of oocytes, the regulatory mechanisms of which have yet to be systematically elucidated. This review discusses the hallmarks of aging to further highlight the main characteristics of ovarian aging and attempt to explore its clinical symptoms and underlying mechanisms. Finally, the intervention strategies related to aging are elaborated, especially the potential role of stem cells and cryopreservation of embryos, oocytes, or ovarian tissue in the delay of ovarian aging.

The ovary is a female reproductive organ that plays a key role in fertility and the maintenance of endocrine homeostasis, which is of great importance to women’s health. It is characterized by a high heterogeneity, with different cellular subpopulations primarily containing oocytes, granulosa cells, stromal cells, endothelial cells, vascular smooth muscle cells, and diverse immune cell types. Each has unique and important functions. From the fetal period to old age, the ovary experiences continuous structural and functional changes, with the gene expression of each cell type undergoing dramatic changes. In addition, ovarian development strongly relies on the communication between germ and somatic cells. Compared to traditional bulk RNA sequencing techniques, the single-cell RNA sequencing (scRNA-seq) approach has substantial advantages in analyzing individual cells within an ever-changing and complicated tissue, classifying them into cell types, characterizing single cells, delineating the cellular developmental trajectory, and studying cell-to-cell interactions. In this review, we present single-cell transcriptome mapping of the ovary, summarize the characteristics of the important constituent cells of the ovary and the critical cellular developmental processes, and describe key signaling pathways for cell-to-cell communication in the ovary, as revealed by scRNA-seq. This review will undoubtedly improve our understanding of the characteristics of ovarian cells and development, thus enabling the identification of novel therapeutic targets for ovarian-related diseases.

Does exogenous calcitonin improve the efficiency of implantation in mice by increasing uterine receptivity? The administration of calcitonin could improve the efficiency of implantation by increasing the expression of several receptivity-related genes in endometrial epithelial cells (EECs). Calcitonin is one of the biomarkers of uterine receptivity, which is transiently produced in the uterine epithelia during the period of implantation both in humans and mouse. Hormone-replaced mice were used for in vivo experiments. To evaluate the effect of calcitonin on uterine receptivity, the expression of endometrial genes was analyzed 36 h after i.p. injection of 0.5 IU calcitonin in a treatment group versus saline in the control. To evaluate the effect of calcitonin on implantation efficiency in vivo, two groups received 0.5 IU or 2 IU calcitonin (i.p.) 24 h before embryo transfer, and a control group received saline (i.p.) (n = 18 mice per group). Implantation sites were counted 7 days after embryo transfer. The RL95-2 human endometrial carcinoma cell line was used to study the mechanisms underlying the effect of calcitonin on gene expression in the endometria. Using an in vitro model of endometrium–trophoblast interaction, established with RL95-2 cells and JAR (human choriocarcinoma cell line) trophoblast, endometrial receptivity was evaluated by comparing attachment and outgrowth of JAR spheroids in control and treatment groups. Uterine receptivity in ovariectomized mice was induced by injection of estradiol and progesterone. Expression of eight genes in murine endometrium and RL95-2 cells was analyzed by real-time RT–PCR, western blot, immunohistochemical analysis, flow cytometry and enzyme-linked immunosorbent assay. We tested the effects of a protein kinase C inhibitor, matrigel and an antibody against integrin αvβ3 using RL95-2 cells and performed attachment and outgrowth assays using the in vitro model of endometrium–trophoblast interaction. Implantation efficiency was evaluated by counting the implantation sites after embryo transfer. Calcitonin up-regulated αvβ3 in RL95 cells, which in turn resulted in increased levels of the leukemia inhibitory factor (LIF) and heparin binding-epidermal growth factor (HB-EGF) mRNA (both P < 0.01 versus control) and protein (both P < 0.05 versus control). The attachment and expansion of JAR spheroids was promoted by pretreatment of EECs with calcitonin (P < 0.05 versus control) together with significantly increased expression of αvβ3, LIF and HB-EGF. Moreover, the injection of calcitonin in the preimplantation phase increased the total number of implantation sites in treatment groups (55 in control versus 78 and 85 in 0.5 and 2 IU groups, respectively). Compared with the control group (3.11 ± 2.14), the average number of implantation sites in the 2 IU calcitonin treatment group increased (4.72 ± 1.87, P = 0.022). Experiments were performed in mice and human cell lines but not in primary cultures of human endometrial cells. The findings presented here have important implications, in that calcitonin administration (currently used for treatment of hypercalcemia or osteoporosis) may have clinical benefits in assisted reproduction programs, by facilitating endometrial receptivity and embryo implantation. However, further studies are required to confirm these findings. This work was supported by National Science Foundation of China (No. 81170619). There are no financial or commercial conflicts in this study.

Premature ovarian insufficiency (POI) is an essential cause of reduced fertility and quality of life in young women. Mesenchymal stem cells (MSCs) and MSCs-derived extracellular vesicles (EVs) have the ability to migrate to damaged tissues and are considered as promising therapeutic approaches for POI. However, the homing ability and therapeutic efficacy of MSCs administered in vivo are still insufficient, and their potential tumorigenicity and multi-differentiation potential also bring many doubts about their safety. The targeting ability and migration efficiency of MSCs can be improved by genetic engineering and surface modification, thereby maximizing their therapeutic efficacy. However, the use of viral vectors also has increased safety concerns. In addition, EVs, which seem to be the current therapeutic alternative to MSCs, are still poorly targeted for distribution, although they have improved in terms of safety. This paper reviews the comparative therapeutic effects of MSCs and their derived EVs on POI, their biodistribution after in vivo administration, and the most important possible ovarian targeting strategies. Difficulties such as homogeneity and yield before clinical application are also discussed. This article will provide new insights into precision therapy and targeted drug delivery for female ovarian diseases. Graphical Abstract