Explore the vast range of titles available.

Visual morphology assessment is routinely used for evaluating of embryo quality and selecting human blastocysts for transfer after in vitro fertilization (IVF). However, the assessment produces different results between embryologists and as a result, the success rate of IVF remains low. To overcome uncertainties in embryo quality, multiple embryos are often implanted resulting in undesired multiple pregnancies and complications. Unlike in other imaging fields, human embryology and IVF have not yet leveraged artificial intelligence (AI) for unbiased, automated embryo assessment. We postulated that an AI approach trained on thousands of embryos can reliably predict embryo quality without human intervention. We implemented an AI approach based on deep neural networks (DNNs) to select highest quality embryos using a large collection of human embryo time-lapse images (about 50,000 images) from a high-volume fertility center in the United States. We developed a framework (STORK) based on Google’s Inception model. STORK predicts blastocyst quality with an AUC of >0.98 and generalizes well to images from other clinics outside the US and outperforms individual embryologists. Using clinical data for 2182 embryos, we created a decision tree to integrate embryo quality and patient age to identify scenarios associated with pregnancy likelihood. Our analysis shows that the chance of pregnancy based on individual embryos varies from 13.8% (age ≥41 and poor-quality) to 66.3% (age <37 and good-quality) depending on automated blastocyst quality assessment and patient age. In conclusion, our AI-driven approach provides a reproducible way to assess embryo quality and uncovers new, potentially personalized strategies to select embryos.

The introduction of time-lapse imaging to clinical in vitro fertilization practice enabled the undisturbed monitoring of embryos throughout the entire culture period. Initially, the main objective was to achieve a better embryo development. However, this technology also provided an insight into the novel concept of morphokinetics, parameters regarding embryo cell dynamics. The vast amount of data obtained defined the optimal ranges in the cell-cycle lengths at different stages of embryo development. This added valuable information to embryo assessment prior to transfer. Kinetic markers became part of embryo evaluation strategies with the potential to increase the chances of clinical success. However, none of them has been established as an international standard. The present work aims at describing new approaches into time-lapse: progress to date, challenges, and possible future directions. Summary Sentence Time-lapse imaging has considerably increased the amount of reliable and detailed information regarding embyo development, contributing to improve the embryo selection.

The success of deep learning in identifying complex patterns exceeding human intuition comes at the cost of interpretability. Non-linear entanglement of image features makes deep learning a “black box” lacking human meaningful explanations for the models’ decision. We present DISCOVER, a generative model designed to discover the underlying visual properties driving image-based classification models. DISCOVER learns disentangled latent representations, where each latent feature encodes a unique classification-driving visual property. This design enables “human-in-the-loop” interpretation by generating disentangled exaggerated counterfactual explanations. We apply DISCOVER to interpret classification of in vitro fertilization embryo morphology quality. We quantitatively and systematically confirm the interpretation of known embryo properties, discover properties without previous explicit measurements, and quantitatively determine and empirically verify the classification decision of specific embryo instances. We show that DISCOVER provides human-interpretable understanding of “black box” classification models, proposes hypotheses to decipher underlying biomedical mechanisms, and provides transparency for the classification of individual predictions. Identifying complex patterns through deep learning often comes at the cost of interpretability. Focusing on the interpretation of classification of in vitro fertilization embryos, the authors present DISCOVER, an approach that enables visual interpretability of image-based classification models.

Assessing fertilized human embryos is crucial for in vitro fertilization, a task being revolutionized by artificial intelligence. Existing models used for embryo quality assessment and ploidy detection could be significantly improved by effectively utilizing time-lapse imaging to identify critical developmental time points for maximizing prediction accuracy. Addressing this, we develop and compare various embryo ploidy status prediction models across distinct embryo development stages. We present BELA, a state-of-the-art ploidy prediction model that surpasses previous image- and video-based models without necessitating input from embryologists. BELA uses multitask learning to predict quality scores that are thereafter used to predict ploidy status. By achieving an area under the receiver operating characteristic curve of 0.76 for discriminating between euploidy and aneuploidy embryos on the Weill Cornell dataset, BELA matches the performance of models trained on embryologists’ manual scores. While not a replacement for preimplantation genetic testing for aneuploidy, BELA exemplifies how such models can streamline the embryo evaluation process. Assessing human embryos is crucial for in vitro fertilization, a task being revolutionized by artificial intelligence. Here, the authors introduce BELA, an automated AI model for predicting embryo ploidy status and quality using time-lapse imaging.

STUDY QUESTIONWhat recommendations can be provided on the approach to and use of time-lapse technology (TLT) in an IVF laboratory?SUMMARY ANSWERThe present ESHRE document provides 11 recommendations on how to introduce TLT in the IVF laboratory.WHAT IS KNOWN ALREADYStudies have been published on the use of TLT in clinical embryology. However, a systematic assessment of how to approach and introduce this technology is currently missing.STUDY DESIGN, SIZE, DURATIONA working group of members of the Steering Committee of the ESHRE Special Interest Group in Embryology and selected ESHRE members was formed in order to write recommendations on the practical aspects of TLT for the IVF laboratory.PARTICIPANTS/MATERIALS, SETTING, METHODSThe working group included 11 members of different nationalities with internationally recognized experience in clinical embryology and basic science embryology, in addition to TLT. This document is developed according to the manual for development of ESHRE recommendations for good practice. Where possible, the statements are supported by studies retrieved from a PUBMED literature search on ‘time-lapse’ and ART.MAIN RESULTS AND THE ROLE OF CHANCEA clear clinical benefit of the use of TLT, i.e. an increase in IVF success rates, remains to be proven. Meanwhile, TLT systems are being introduced in IVF laboratories. The working group listed 11 recommendations on what to do before introducing TLT in the lab. These statements include an assessment of the pros and cons of acquiring a TLT system, selection of relevant morphokinetic parameters, selection of an appropriate TLT system with technical and customer support, development of an internal checklist and education of staff. All these aspects are explained further here, based on the current literature and expert opinion.LIMITATIONS, REASONS FOR CAUTIONOwing to the limited evidence available, recommendations are mostly based on clinical and technical expertise. The paper provides technical advice, but leaves any decision on whether or not to use TLT to the individual centres.WIDER IMPLICATIONS OF THE FINDINGSThis document is expected to have a significant impact on future developments of clinical embryology, considering the increasing role and impact of TLT.STUDY FUNDING/COMPETING INTEREST(S)The meetings of the working group were funded by ESHRE. S.A. declares participation in the Nordic Embryology Academic Team with meetings sponsored by Gedeon Richter. T.E. declares to have organized workshops for Esco and receiving consulting fees from Ferring and Gynemed and speakers’ fees from Esco and honorarium from Merck and MSD. T.F. received consulting fees from Vitrolife and Laboratoires Genévrier, speakers’ fees from Merck Serono, Gedeon Richter, MSD and Ferring and research grants from Gedeon Richter and MSD. M.M. received sponsorship from Merck. M.M.E. received speakers’ fees from Merck, Ferring and MSD. R.S. received a research grant from ESHRE. G.C. received speakers’ fees from IBSA and Excemed. The other authors declare that they have no conflict of interest.TRIAL REGISTRATION NUMBERN/A.DISCLAIMER This Good Practice Recommendations (GPR) document represents the views of ESHRE, which are the result of consensus between the relevant ESHRE stakeholders and are based on the scientific evidence available at the time of preparation. ESHRE’s GPRs should be used for information and educational purposes. They should not be interpreted as setting a standard of care or be deemed inclusive of all proper methods of care nor exclusive of other methods of care reasonably directed to obtaining the same results. They do not replace the need for application of clinical judgment to each individual presentation, nor variations based on locality and facility type. Furthermore, ESHRE GPRs do not constitute or imply the endorsement, or favouring of any of the included technologies by ESHRE. †ESHRE Pages content is not externally peer reviewed. The manuscript has been approved by the Executive Committee of ESHRE.

Embryo assessment in in vitro fertilization (IVF) involves multiple tasks—including ploidy prediction, quality scoring, component segmentation, embryo identification, and timing of developmental milestones. Existing methods address these tasks individually, leading to inefficiencies due to high costs and lack of standardization. Here, we introduce FEMI (Foundational IVF Model for Imaging), a foundation model trained on approximately 18 million time-lapse embryo images. We evaluate FEMI on ploidy prediction, blastocyst quality scoring, embryo component segmentation, embryo witnessing, blastulation time prediction, and stage prediction. FEMI attains area under the receiver operating characteristic (AUROC) >  0.75 for ploidy prediction using only image data—significantly outpacing benchmark models. It has higher accuracy than both traditional and deep-learning approaches for overall blastocyst quality and its subcomponents. Moreover, FEMI has strong performance in embryo witnessing, blastulation-time, and stage prediction. Our results demonstrate that FEMI can leverage large-scale, unlabelled data to improve predictive accuracy in several embryology-related tasks in IVF. In vitro fertilisation relies on accurate, non-invasive embryo evaluation to improve success rates. Here, authors present a foundation model trained on 18 million time-lapse images, which outperforms existing benchmarks in ploidy prediction, quality scoring, segmentation, and developmental timing.

Background The application of Artificial Intelligence (AI) to sperm selection during Intracytoplasmic Sperm Injection (ICSI) procedures represents one of the most innovative advances in assisted reproductive technology (ART). Traditional sperm selection relies heavily on the subjective assessment of embryologists, which can lead to variability in outcomes. This study aimed to evaluate the performance of an AI-based software, Sperm ID (SiD™) v.1.0, for sperm selection during ICSI and to compare its outcomes with those obtained by experienced embryologists. Additionally, the study assessed the potential impact of sperm and oocyte quality, particularly in autologous versus donor oocyte cycles. Methods A single-center, blind, observational study was conducted involving 102 infertile couples—60 undergoing treatment with autologous oocytes and 42 using oocytes from a donation program. Semen samples were analyzed in real time with SiD™ v.1.0, a software that quantifies progressive motility parameters and assigns each sperm a categorical score (‘Best,’ ‘Good,’ ‘Medium,’ or ‘low’). Spermatozoa and oocytes were individually tracked from injection to embryo development. Oocyte quality was retrospectively analyzed using another AI tool, Magenta IVF R3.0. The performance of the Artificial Intelligence Sperm Selection (AISS) system was compared with that of senior embryologists (> 300 ICSI cycles/year). Statistical analysis included descriptive statistics and inferential tests to compare fertilization and embryo development rates across sperm categories and between autologous and donor cycles. Results Biological outcomes—such as fertilization and blastocyst development—were generally similar across all sperm quality categories. However, in cycles with autologous oocytes, the use of top-quality sperm (‘Best’ category) was associated with a significantly higher blastocyst formation rate. In contrast, no significant differences were observed in donor oocyte cycles, regardless of sperm quality. The AISS system demonstrated comparable performance to that of senior embryologists, with similar fertilization and embryo development rates. Conclusions The study highlights the promising role of AI-based tools in standardizing and enhancing sperm selection during ICSI. While AI-driven sperm selection showed limited impact in donor cycles, it may offer a distinct advantage in cases involving compromised oocyte quality. Furthermore, AISS may improve laboratory efficiency and support junior embryologists by reducing selection time and increasing procedural consistency.

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.

Spontaneous blastocyst collapse during in vitro embryo development has been suggested as a novel marker of embryo quality. Therefore, the aim of this multicentre study was to carry out a retrospective multicentre analysis to investigate the correlation between blastocyst collapse and pregnancy outcome. Here, 1297 intracytoplasmic sperm injection (ICSI)/ in vitro fertilization (IVF) fresh cycles, with an elective single blastocyst transfer (eSET) were included in this study. Embryos were cultured individually in 6.0% CO 2 , 5.0% O 2 , 89.0% N 2 , using single step medium (GTL TM VitroLife, Sweden) or sequential medium (Cook TM , Cook Medical, Australia) and selected for transfer using standard morphological criteria. With the use of time-lapse monitoring (TLM), blastocysts were analyzed by measuring the maximum volume reduction and defined as having collapsed, if there was ≥ 50% volume reduction from the expanded blastocyst and the collapse event. Following embryo replacement, each blastocyst was retrospectively allocated to one of two groups (collapsed or not collapsed). Here, 259 blastocysts collapsed once or more during development (19.9%) and the remaining 1038 either contracted minimally or not collapsed (80.1%). A significantly higher ongoing pregnancy rate (OPR) of 51.9% (95% CI 48.9–59.9%) was observed when blastocysts that had not collapsed were replaced compared with cycles in which collapsed blastocysts were transferred 37.5% (95% CI 31.6–43.4%). This study suggests that human blastocysts that collapse spontaneously during development are less likely to implant and generate a pregnancy compared with embryos that do not. Although this is a retrospective study, the results demonstrated the utility of collapse episodes as new marker of embryo selection following eSET at blastocyst stage.

Background: Sperm selection is critical in assisted reproduction, typically relying on swim-up and centrifugation density gradients. New methods, such as PICSI (physiological intracytoplasmic sperm selection), aim to enhance outcomes by selecting mature sperm based on hyaluronic acid (HA) binding and have generated interest due to their potential impact on the clinical outcomes of patients who undergo assisted reproductive treatments. Methods: A single-center, prospective, and triple-blinded study was conducted with 277 couples in the egg donation program. The oocytes of each recipient patient were randomly microinjected using the ICSI or PICSI technique and maintained in culture in time-lapse incubators until blastocyst formation. Biological and clinical outcomes were analyzed, including fertilization and blastocyst formation rates, embryo morphokinetics, pregnancy, miscarriage, and live birth rates, and artificial intelligence-assigned embryo quality scores. Results: Clinical outcomes were comparable between the two groups, but a higher pregnancy rate was observed in the PICSI group than in the ICSI group (74.04% vs. 70.87%). Although blastocyst formation rates were similar on both day 5 (D5) and day 6 of development, the proportion of good-quality embryos on D5 was higher in the PICSI group (68.27%) than in the ICSI group (63.47%) (p > 0.05). Finally, the cumulative pregnancy rate was significantly higher in the experimental group than in the control group (88% vs. 72%) after four embryo transfers (p < 0.05). Conclusions: Utilizing HA to perform sperm selection during ICSI procedures does not increase live birth rates. However, it may enhance the quality of the selected sperm. This could be beneficial for patients in egg donation programs, particularly for those who have experienced repeated pregnancy loss.