Search

Study question Are the annotations produced by AI comparable to manual annotations? Does AI accurately assess fertilisation checks, and predict embryo usage and blastulation compared to embryologists? Summary answer Automatic annotations by AI was consistent with manual annotations. AI implantation algorithms had strong prediction of blastulation and embryo usage. What is known already Currently, embryos are manually annotated for specific morphokinetic features during embryo development. This is a labour-intensive process, and dependent on training and experience, leading to inter and intra clinic variation. The decision to transfer, freeze or discard embryos relies heavily on these annotations. It is paramount that we develop a tool that will provide consistency and accuracy in annotation and produce scores that can facilitate decisions around embryo usage. AI has demonstrated its potential to achieve this, but first must be validated before its integration into clinical practice. There have been no such studies demonstrating this so far. Study design, size, duration Retrospective cohort study, that took place between September to December 2021 at a private fertility clinic in Spain. To control for embryo variability, this study only included 179 time-lapse videos for embryos created from donor eggs. This was based on the understanding that donor eggs are more likely to produce better quality blastocysts and embryos and thus will give the most optimal conditions for annotation in a validation framework. Participants/materials, setting, methods The same time-lapse cultured embryos were annotated manually and automatically by CHLOE(Fairtility, an AI-based tool). Manual and CHLOE annotations were compared to assess the strength of agreement (i) using intra-class correlation (ICC), and (ii) the proportion of corrections required at the pronuclei (PN) stage. AI accuracy in predicting blastulation at 30hours, and blastulation before 116 hours, was also assessed using AUC as the efficacy metric. Embryo usage was compared with the AI-generated ranking of embryos. Main results and the role of chance The majority of morphokinetic variables showed a very-strong agreement, with an ICC range of (0.81-1.00), namely for; tPNf, t2, t3, t5, t7, tSB, tB and tEB. Only t4 (0.5) showed a moderate agreement. On average (Mean+-Standard deviation), AI annotated t4 later than embryologists (36+-5vs39+-10 (hours)). All other variables fell within a strong ICC of (0.61-0.8). There were no very weak (0-0.2) or weak (0.21-0.4) variables. PN agreement between AI and embryologists was 93%: PN’s had to be corrected by an embryologist only 7%(n = 179) of the time. AI predicted blastulation on day 3 with a high level of sensitivity 0.77 and specificity 0.82, (AUC: 0.84,p Similarly, AI-generated ranking accurately correlated with embryologist decisions to freeze, transfer or discard embryos, with an overall high sensitivity 0.88 and specificity 0.67, (AUC: 0.84,p Limitations, reasons for caution This study only included embryos from donor eggs. Furthermore, this study occurred at a single site and is planned to be replicated at several clinics. Where there are discrepancies between human and AI, further studies are required to determine the ground truth. Wider implications of the findings This study demonstrates an AI framework to safely introduce AI in the fertility clinic. AI will accurately annotate embryos and give reliable scores to predict good quality blastulation, and inform decisions around embryo usage determination. AI provides a time-effective, objective tool in decision-making, with the potential to optimise success. Trial registration number not applicable

Study question Can lab-related variables (media type, oil viscosity, microdroplet volume and culture dish design) modulate media evaporation and improve its stability during culture? Summary answer Using dishes with pre-defined wells, big volume microdroplets and high-viscosity mineral oil can help to reduce media evaporation and improve osmolality stability during embryo culture. What is known already Osmolality measures the number of solute particles present in a solution and is an important variable of a human embryo culture system. High ambient temperature and low humidity may induce evaporation in culture media, increasing its osmolality. In addition, recent tendencies in IVF laboratories, such as extending the embryo culture uninterruptedly until day 6/7 or the use of dry benchtop incubators, may intensify evaporation. Surpassing a 300mOsm/kg threshold can result deleterious for embryo development and impair clinical results. Different strategies (e.g. oil type/volume, dish type, micro-drop volume) have been proposed to reduce evaporation and stabilize osmolality during culture. Study design, size, duration Four variables were analyzed in their capacity to reduce media evaporation: type of culture medium, micro-droplet volume, oil viscosity and type of culture dish. Dishes were prepared with 5ml of oil and 50µl microdroplets (25µl were used for the comparison of micro-droplet volumes). Dishes were cultured in parallel in a dry benchtop incubator (AD–3100, Astec), and osmolality measured daily for seven days with a freezing point depression osmometer (Osmo1®, Advanced Instruments, accuracy ≤2mOsm/kg). Participants/materials, setting, methods The following comparison groups were analyzed: 1) Seven commercial single-step media with three differing initial osmolalities (approximately 260, 280 and >290mOsm/kg); 2) oil with high, medium or low viscosity; 3) 50 vs. 25µl microdroplets; 4) 35mm flat Petri dish vs. 35mm dish with defined wells. Temperature in the incubator was monitored continuously (T+Button, BrightSentinel), as well as room temperature and humidity (Octax Log&Guard, Vitrolife). All were stable at 37.3±0.05oC, 22.1±0.6 oC and 67.4±7.4%, respectively. Main results and the role of chance Evaporation occurred in all the studied groups, but its rate was modulated by various parameters. Culture dishes designed with pre-defined wells reduced evaporation when compared to regular Petri dishes (Increase 11.3mOsm/kg and increase 12.5mOsm/kg, respectively from day 0 to 7 (P = 0.007)). Similarly, oil viscosity had an impact in osmolality stability during culture, with an increase of 14.7mOsm/kg, 16.3mOsm/kg and 19.2mOsm/kg observed when using mineral oil with high, medium and low viscosity, respectively (P = 0.009). Finally, reducing the volume of the medium microdroplets from 50 to 25µl derived in higher evaporation rates, but without significant differences (Increase 14.7mOsm/kg and increase 15.8mOsm/kg, respectively (P = 0.325)). Different evaporation rates were observed between the seven studied culture media attending their three-differing initial osmolalities. Significant differences were observed for a media respect another three media with differing initial osmolality (P = 0.001, P = 0.01 and P = 0.015). Their initial osmolality had a direct correlation with the maximum osmolality reached at the end of culture. Thus, media with a high initial osmolality (>290mOsm/kg) resulted in hyperosmotic media above the recommended 300mOsm/kg threshold by the end of culture and, by contrast, the studied media with lower initial values were able to maintain osmolality below 300mOsm/kg for the whole duration of the culture. Limitations, reasons for caution While a clear effect was observed by the studied variables, other parameters, such as oil volume or dish preparation techniques, could also play a role in osmolality maintenance and could be studied in the future. Additionally, these findings could vary between different centers and should be validated in each laboratory. Wider implications of the findings: Osmolality has been shown to have a direct impact on embryo development, embryo quality and clinical outcomes. Carefully defining the consumables and methodologies used in the IVF laboratory will improve the stability of the culture system and, consequently, reduce the stress imparted to the embryos and gametes under culture. Trial registration number Not applicable