An artificial intelligence model correlated with morphological and genetic features of blastocyst quality improves ranking of viable embryos.

RESEARCH QUESTION: Can better methods be developed to evaluate the performance and characteristics of an artificial intelligence model for evaluating the likelihood of clinical pregnancy based on analysis of day-5 blastocyst-stage embryos, such that performance evaluation more closely reflects clinical use in IVF procedures, and correlations with known features of embryo quality are identified? DESIGN: De-identified images were provided retrospectively or collected prospectively by IVF clinics using the artificial intelligence model in clinical practice. A total of 9359 images were provided by 18 IVF clinics across six countries, from 4709 women who underwent IVF between 2011 and 2021. Main outcome measures included clinical pregnancy outcome (fetal heartbeat at first ultrasound scan), embryo morphology score, and/or pre-implantation genetic testing for aneuploidy (PGT-A) results. RESULTS: A positive linear correlation of artificial intelligence scores with pregnancy outcomes was found, and up to a 12.2% reduction in time to pregnancy (TTP) was observed when comparing the artificial intelligence model with standard morphological grading methods using a novel simulated cohort ranking method. Artificial intelligence scores were significantly correlated with known morphological features of embryo quality based on the Gardner score, and with previously unknown morphological features associated with embryo ploidy status, including chromosomal abnormalities indicative of severity when considering embryos for transfer during IVF. CONCLUSION: Improved methods for evaluating artificial intelligence for embryo selection were developed, and advantages of the artificial intelligence model over current grading approaches were highlighted, strongly supporting the use of the artificial intelligence model in a clinical setting.

Tracking intracellular forces and mechanical property changes in mouse one-cell embryo development.

Cells comprise mechanically active matter that governs their functionality, but intracellular mechanics are difficult to study directly and are poorly understood. However, injected nanodevices open up opportunities to analyse intracellular mechanobiology. Here, we identify a programme of forces and changes to the cytoplasmic mechanical properties required for mouse embryo development from fertilization to the first cell division. Injected, fully internalized nanodevices responded to sperm decondensation and recondensation, and subsequent device behaviour suggested a model for pronuclear convergence based on a gradient of effective cytoplasmic stiffness. The nanodevices reported reduced cytoplasmic mechanical activity during chromosome alignment and indicated that cytoplasmic stiffening occurred during embryo elongation, followed by rapid cytoplasmic softening during cytokinesis (cell division). Forces greater than those inside muscle cells were detected within embryos. These results suggest that intracellular forces are part of a concerted programme that is necessary for development at the origin of a new embryonic life.

AI in the treatment of fertility: key considerations.

Artificial intelligence (AI) has been proposed as a potential tool to help address many of the existing problems related with empirical or subjective assessments of clinical and embryological decision points during the treatment of infertility. AI technologies are reviewed and potential areas of implementation of algorithms are discussed, highlighting the importance of following a proper path for the development and validation of algorithms, including regulatory requirements, and the need for ecosystems containing enough quality data to generate it. As evidenced by the consensus of a group of experts in fertility if properly developed, it is believed that AI algorithms may help practitioners from around the globe to standardize, automate, and improve IVF outcomes for the benefit of patients. Collaboration is required between AI developers and healthcare professionals to make this happen.

Transcriptome asymmetry within mouse zygotes but not between early embryonic sister blastomeres.

Transcriptome regionalization is an essential polarity determinant among metazoans, directing embryonic axis formation during normal development. Although conservation of this principle in mammals is assumed, recent evidence is conflicting and it is not known whether transcriptome asymmetries exist within unfertilized mammalian eggs or between the respective cleavage products of early embryonic divisions. We here address this by comparing transcriptome profiles of paired single cells and sub-cellular structures obtained microsurgically from mouse oocytes and totipotent embryos. Paired microsurgical spindle and remnant samples from unfertilized metaphase II oocytes possessed distinguishable profiles. Fertilization produces a totipotent 1-cell embryo (zygote) and associated spindle-enriched second polar body whose paired profiles also differed, reflecting spindle transcript enrichment. However, there was no programmed transcriptome asymmetry between sister cells within 2- or 3-cell embryos. Accordingly, there is transcriptome asymmetry within mouse oocytes, but not between the sister blastomeres of early embryos. This work places constraints on pre-patterning in mammals and provides documentation correlating potency changes and transcriptome partitioning at the single-cell level.

Epigenetic characterization of the early embryo with a chromatin immunoprecipitation protocol applicable to small cell populations.

Chromatin immunoprecipitation (ChIP) defines the genomic distribution of proteins and their modifications but is limited by the cell numbers required (ideally >10(7)). Here we describe a protocol that uses carrier chromatin and PCR, 'carrier' ChIP (CChIP), to permit analysis of as few as 100 cells. We assayed histone modifications at key regulator genes (such as Nanog, Pou5f1 (also known as Oct4) and Cdx2) by CChIP in mouse embryonic stem (ES) cells and in inner cell mass (ICM) and trophectoderm of cultured blastocysts. Activating and silencing modifications (H4 acetylation and H3K9 methylation) mark active and silent promoters as predicted, and we find close correlation between values derived from CChIP (1,000 ES cells) and conventional ChIP (5 x 10(7) ES cells). Studies on genes silenced in both ICM and ES cells (Cdx2, Cfc1, Hhex and Nkx2-2, also known as Nkx) show that the intensity of silencing marks is relatively diminished in ES cells, indicating a possible relaxation of some components of silencing on adaptation to culture.

Computer-automated time-lapse analysis results correlate with embryo implantation and clinical pregnancy: a blinded, multi-centre study.

Computer-automated time-lapse analysis has been shown to improve embryo selection by providing quantitative and objective information to supplement traditional morphology. In this multi-centre study, the relationship between such computer-derived outputs (High, Medium, Low scores), embryo implantation and clinical pregnancy were examined. Data were collected from six clinics, including 205 patients whose embryos were imaged by the Eeva(TM) System. The Eeva scores were blinded and not considered during embryo selection. Embryos with High and Medium scores had significantly higher implantation rates than those with Low scores (37% and 35% versus 15%; P < 0.0001; P = 0.0004). Similar trends in implantation rates were observed in different IVF centres each using their own protocols. Further analysis revealed that patients with at least one High embryo transferred had significantly higher clinical pregnancy rates than those with only Low embryos transferred (51% versus 34%; P = 0.02), although patients' clinical characteristics across groups were comparable. These data, together with previous research and clinical studies, confirm that computer-automated Eeva scores provide valuable information, which may improve the clinical outcome of IVF procedures and ultimately facilitate the trend of single embryo selection.

A program of successive gene expression in mouse one-cell embryos.

At the moment of union in fertilization, sperm and oocyte are transcriptionally silent. The ensuing onset of embryonic transcription (embryonic genome activation [EGA]) is critical for development, yet its timing and profile remain elusive in any vertebrate species. We here dissect transcription during EGA by high-resolution single-cell RNA sequencing of precisely synchronized mouse one-cell embryos. This reveals a program of embryonic gene expression (immediate EGA [iEGA]) initiating within 4 h of fertilization. Expression during iEGA produces canonically spliced transcripts, occurs substantially from the maternal genome, and is mostly downregulated at the two-cell stage. Transcribed genes predict regulation by transcription factors (TFs) associated with cancer, including c-Myc. Blocking c-Myc or other predicted regulatory TF activities disrupts iEGA and induces acute developmental arrest. These findings illuminate intracellular mechanisms that regulate the onset of mammalian development and hold promise for the study of cancer.

An assessment tool for computer-assisted semen analysis (CASA) algorithms.

Computer-Assisted Semen Analysis (CASA) enables reliable analysis of semen images, and is designed to process large number of images with high consistency, accuracy, and repeatability. Design and testing of CASA algorithms can be accelerated greatly if reliable simulations of semen images under a variety of conditions and sample quality modes are available. Using life-like simulation of semen images can quantify the performance of existing and proposed CASA algorithms, since the parameters of the simulated image are known and controllable. We present simulation models for sperm cell image and swimming modes observed in real 2D (top-down) images of sperm cells in laboratory specimen. The models simulate human sperm using four (4) types of swimming, namely linear mean, circular, hyperactive, and immotile (or dead). The simulation models are used in studying algorithms for segmentation, localization, and tracking of sperm cells. Several segmentation and localization algorithms were tested under varying levels of noise, and then compared using precision, recall, and the optimal subpattern assignment (OSPA) metric. Images of real human semen sample were used to validate the segmentation and localization observations obtained from simulations. An example is given of sperm cell tracking on simulated semen images of cells using the different tracking algorithms (nearest neighbor (NN), global nearest neighbor (GNN), probabilistic data association filter (PDAF), and joint probabilistic data association filter (JPDAF)). Tracking performance was evaluated through multi-object tracking precision (MOTP) and multi-object tracking accuracy (MOTA). Simulation models enable objective assessments of semen image processing algorithms. We demonstrate the use of a new simulation tool to assess and compare segmentation, localization, and tracking methods. The simulation software allows testing along a large spectrum of parameter values that control the appearance and behavior of simulated semen images. Users can generate scenarios of different characteristics and assess the effectiveness of different CASA algorithms in these environments. The simulation was used to assess and compare algorithms for segmentation and tracking of sperm cells in semen images.

Characterization of an artificial intelligence model for ranking static images of blastocyst stage embryos.

OBJECTIVE: To perform a series of analyses characterizing an artificial intelligence (AI) model for ranking blastocyst-stage embryos. The primary objective was to evaluate the benefit of the model for predicting clinical pregnancy, whereas the secondary objective was to identify limitations that may impact clinical use. DESIGN: Retrospective study. SETTING: Consortium of 11 assisted reproductive technology centers in the United States. PATIENT(S): Static images of 5,923 transferred blastocysts and 2,614 nontransferred aneuploid blastocysts. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Prediction of clinical pregnancy (fetal heartbeat). RESULT(S): The area under the curve of the AI model ranged from 0.6 to 0.7 and outperformed manual morphology grading overall and on a per-site basis. A bootstrapped study predicted improved pregnancy rates between +5% and +12% per site using AI compared with manual grading using an inverted microscope. One site that used a low-magnification stereo zoom microscope did not show predicted improvement with the AI. Visualization techniques and attribution algorithms revealed that the features learned by the AI model largely overlap with the features of manual grading systems. Two sources of bias relating to the type of microscope and presence of embryo holding micropipettes were identified and mitigated. The analysis of AI scores in relation to pregnancy rates showed that score differences of ≥0.1 (10%) correspond with improved pregnancy rates, whereas score differences of <0.1 may not be clinically meaningful. CONCLUSION(S): This study demonstrates the potential of AI for ranking blastocyst stage embryos and highlights potential limitations related to image quality, bias, and granularity of scores.

Human embryonic genome activation initiates at the one-cell stage.

In human embryos, the initiation of transcription (embryonic genome activation [EGA]) occurs by the eight-cell stage, but its exact timing and profile are unclear. To address this, we profiled gene expression at depth in human metaphase II oocytes and bipronuclear (2PN) one-cell embryos. High-resolution single-cell RNA sequencing revealed previously inaccessible oocyte-to-embryo gene expression changes. This confirmed transcript depletion following fertilization (maternal RNA degradation) but also uncovered low-magnitude upregulation of hundreds of spliced transcripts. Gene expression analysis predicted embryonic processes including cell-cycle progression and chromosome maintenance as well as transcriptional activators that included cancer-associated gene regulators. Transcription was disrupted in abnormal monopronuclear (1PN) and tripronuclear (3PN) one-cell embryos. These findings indicate that human embryonic transcription initiates at the one-cell stage, sooner than previously thought. The pattern of gene upregulation promises to illuminate processes involved at the onset of human development, with implications for epigenetic inheritance, stem-cell-derived embryos, and cancer.

Dosage compensation in the mouse balances up-regulation and silencing of X-linked genes.

Dosage compensation in mammals involves silencing of one X chromosome in XX females and requires expression, in cis, of Xist RNA. The X to be inactivated is randomly chosen in cells of the inner cell mass (ICM) at the blastocyst stage of development. Embryonic stem (ES) cells derived from the ICM of female mice have two active X chromosomes, one of which is inactivated as the cells differentiate in culture, providing a powerful model system to study the dynamics of X inactivation. Using microarrays to assay expression of X-linked genes in undifferentiated female and male mouse ES cells, we detect global up-regulation of expression (1.4- to 1.6-fold) from the active X chromosomes, relative to autosomes. We show a similar up-regulation in ICM from male blastocysts grown in culture. In male ES cells, up-regulation reaches 2-fold after 2-3 weeks of differentiation, thereby balancing expression between the single X and the diploid autosomes. We show that silencing of X-linked genes in female ES cells occurs on a gene-by-gene basis throughout differentiation, with some genes inactivating early, others late, and some escaping altogether. Surprisingly, by allele-specific analysis in hybrid ES cells, we also identified a subgroup of genes that are silenced in undifferentiated cells. We propose that X-linked genes are silenced in female ES cells by spreading of Xist RNA through the X chromosome territory as the cells differentiate, with silencing times for individual genes dependent on their proximity to the Xist locus.

Appendix D: Irvine Scientific® Vitrification System.

This chapter will describe the use of the Irvine Scientific® vitrification and warming solutions (Vit Kits®) along with a detailed protocol for the correct use of the CryoTip® vitrification device for human embryos and oocytes. Successful pregnancies have been reported after carrying out rapid vitrification methods of oocytes, cleavage stage embryos, and blastocysts using the CryoTip® (Popwell et al. Fertil Steril, 101:e20, 2014; Kuwayama et al. Reprod Biomed 11:608-614, 2005; Kuwayama et al. Fertil Steril 84:S187, 2005; Kuwayama et al. Vitrification of human embryos using the CryoTip TM method. Reprod Biomed 11:608-614). Compared to other vitrification carrier devices, the CryoTip® is considered a closed carrier for vitrification, thereby eliminating the theoretical risk of disease transmission through contaminated liquid nitrogen during cooling and storage (Bielanski et al. Cryobiology, 40:110-116, 2000). The CryoTip® is cleared by the FDA and has CE mark approval as a closed device to carry gametes or embryos in a specialized medium during cryopreservation procedures and subsequent long-term storage in liquid nitrogen. The CryoTip® has been shown to be a safe and reliable vitrification device, and when compared to other open system vitrification devices, it has provided similar results (Martino et al. Reprod Biol Endocrinol 11:27, 2013, Valbuena et al. Fertil Steril 97:209-217, 2012; Kuwayama et al. Reprod Biomed 11:608-614, 2005). The CryoTip® has also been shown to be suitable for use as a vitrification device for the cryopreservation of small volumes of sperm (Tanaka et al. Fertil Steril 90: S292, 2008).

Automatic Tracking and Motility Analysis of Human Sperm in Time-Lapse Images.

We present a fully automated multi-sperm tracking algorithm. It has the demonstrated capability to detect and track simultaneously hundreds of sperm cells in recorded videos while accurately measuring motility parameters over time and with minimal operator intervention. Algorithms of this kind may help in associating dynamic swimming parameters of human sperm cells with fertility and fertilization rates. Specifically, we offer an image processing method, based on radar tracking algorithms, that detects and tracks automatically the swimming paths of human sperm cells in timelapse microscopy image sequences of the kind that is analyzed by fertility clinics. Adapting the well-known joint probabilistic data association filter (JPDAF), we automatically tracked hundreds of human sperm simultaneously and measured their dynamic swimming parameters over time. Unlike existing CASA instruments, our algorithm has the capability to track sperm swimming in close proximity to each other and during apparent cell-to-cell collisions. Collecting continuously parameters for each sperm tracked without sample dilution (currently impossible using standard CASA systems) provides an opportunity to compare such data with standard fertility rates. The use of our algorithm thus has the potential to free the clinician from having to rely on elaborate motility measurements obtained manually by technicians, speed up semen processing, and provide medical practitioners and researchers with more useful data than are currently available.

Transcription-independent heritability of induced histone modifications in the mouse preimplantation embryo.

Enzyme-catalyzed, post-translational modifications of core histones have been implicated in the complex changes in gene expression that drive early mammalian development. However, until recently the small number of cells available from the preimplantation embryo itself has prevented quantitative analysis of histone modifications at key regulator genes. The possible involvement of histone modifications in the embryo's response to extracellular signals, or as determinants of cell fate or lineage progression, remains unclear. Here we describe the use of a recently-developed chromatin immunoprecipitation technique (CChIP) to assay histone modification levels at key regulator genes (Pou5f1, Nanog, Cdx2, Hoxb1, Hoxb9) as mouse embryos progress from 8-cell to blastocyst in culture. Only by the blastocyst stage, when the embryonic (Inner Cell Mass) and extra-embryonic (Trophoblast) lineages are compared, do we see the expected association between histone modifications previously linked to active and silent chromatin, and transcriptional state. To explore responses to an environmental signal, we exposed embryos to the histone deacetylase inhibitor, anti-epileptic and known teratogen valproic acid (VPA), during progression from 8-cell to morula stage. Such treatment increased H4 acetylation and H3 lysine 4 methylation at the promoters of Hoxb1 and Hoxb9, but not the promoters of Pou5f1, Nanog,Cdx2 or the housekeeping gene Gapdh. Despite the absence of detectable Hoxb transcription, these VPA-induced changes were heritable, following removal of the inhibitor, at least until the blastocyst stage. The selective hyperacetylation of Hoxb promoters in response to a histone deacetylase inhibitor, suggests that Hox genes have a higher turnover of histone acetates than other genes in the preimplantation embryo. To explain the heritability, through mitosis, of VPA-induced changes in histone modification at Hoxb promoters, we describe how an epigenetic feed-forward loop, based on cross-talk between H3 acetylation and H3K4 methylation, might generate a persistently increased steady-state level of histone acetylation in response to a transient signal.