Morphokinetic Parameters Research Articles

Cleavage-Stage Embryo Segmentation Using SAM-Based Dual Branch Pipeline: Development and Evaluation with the CleavageEmbryo Dataset.

Embryo selection is one of the critical factors in determining the success of pregnancy in in vitro fertilization (IVF) procedures. Using artificial intelligence to aid in embryo selection could effectively address the current time-consuming, expensive, subjectively influenced process of embryo assessment by trained embryologists. However, current deep learning-based methods often focus on blastocyst segmentation, grading, or predicting cell development via time-lapse videos, often overlooking morphokinetic parameters or lacking interpretability. Given the significance of both morphokinetic and morphological evaluation in predicting the implantation potential of cleavage-stage embryos, as emphasized by previous research, there is a necessity for an automated method to segment cleavage-stage embryos to improve this process. In this article, we introduce the SAM-based Dual Branch Segmentation Pipeline for automated segmentation of blastomeres in cleavage-stage embryos. Leveraging the powerful segmentation capability of SAM, the instance branch conducts instance segmentation of blastomeres, while the semantic branch performs semantic segmentation of fragments. Due to the lack of publicly available datasets, we construct the CleavageEmbryo dataset, the first dataset of human cleavage-stage embryos with pixel-level annotations containing fragment information. We train and test a series of state-of-the-art segmentation algorithms on CleavageEmbryo. Our experiments demonstrate that our method outperforms existing algorithms in terms of objective metrics (mAP 0.748 on blastomeres, Dice 0.694 on fragments) and visual quality, enabling more accurate segmentation of cleavage-stage embryos. The code and sample data in this study can be found at: Https://github.com/12austincc/Cleavage-StageEmbryoSegmentation. Supplementary data are available at Bioinformatics online.

The kinetics of nucleolar precursor bodies clustering at the pronuclei interface: Positive correlations with the morphokinetic characteristics of cleaving embryos and euploidy in preimplantation genetic testing programs

Objective: This study investigated potential relationships between the kinetics of nucleolar precursor bodies (NPBs) in the pronucleus and developmental morphokinetics and euploidy in human preimplantation genetic testing for aneuploidy (PGT-A) cycles.Methods: The morphokinetic analysis of 200 blastocysts obtained from 53 PGT-A cycles was performed retrospectively in a time-lapse incubator. At the time of pronuclear breakdown (PNBD), we categorized the blastocysts into two groups based on the kinetic degree of clustering NPBs at the interface of the two pronuclei: clustered NPBs (CL) and non-clustered NPBs (NCL). We then compared morphokinetic parameters, abnormal behavioral events, and the rate of aneuploidy between the two groups.Results: Pronuclear fading and the first cleavage occurred earlier in the NCL group than in the CL group. However, the initiation of blastocyst formation and blastocyst expansion was delayed in the NCL group relative to the CL group. No differences were found in the rate of abnormal cleavage events, such as multinucleation at the 2-cell stage, direct cleavage from one to three cells, and from two to five cells between the CL and NCL groups. However, the fragmentation rate at the 8-cell stage was higher in the NCL group than in the CL group (10.3% vs. 1.9%, p<0.05). Additionally, the euploid rate in the CL group was significantly higher than in the NCL group (37.9% vs. 12.4%, p<0.05).Conclusion: These results demonstrate the effectiveness of combining NPB clustering at PNBD with morphokinetics as a parameter for selecting embryos with higher developmental potential in in vitro fertilization.

Correlation between Human Embryo Morphokinetics Observed through Time-Lapse Incubator and Life Birth Rate.

(1) Background: Does the variation in sequential development times of embryos, observed through time-lapse monitoring, between the two study groups play a role in predicting pregnancy success? (2) Methods: The prospective double-arm study was to identify the morphokinetic parameters specific to embryos that were capable of implanting and were conducted on 89 embryos cultured in the Esco Miri time-lapse incubator, divided into two groups: Lot A, consisting of 57 embryos that successfully implanted and resulted in life birth rate (LBR), and Lot B (NLB), comprising 32 embryos that did not implant, leading to a negative beta-hCG outcome. (3) Results: Baseline characteristics, including female age, were not found to be statistically significant (p > 0.01). In contrast, there is a highly statistically significant difference concerning oocytes (p = 0.0029). Morphokinetic variables represented by sequential culture times were not statistically significant (p > 0.01) when comparing the two groups. However, the negative mean differences between these parameters suggest that the times for Lot A are better (shorter) than those for Lot B. While not statistically significant, these differences may still have practical significance. In the case of grading, the difference is considered to be extremely statistically significant (p < 0.01). (4) Conclusions: Although there are no statistically significant differences in sequential timings (p > 0.01) between the two groups, there are parameters indicating predictive potential for exploring pregnancy in embryo morphokinetics.

The effect of male factors on embryo morphokinetics: a retrospective analysis of 2726 blastocysts.

Male infertility may influence fertilization rates, embryo morphology, and implantation rates in in vitro fertilization (IVF) cycles. Oocyte competence plays a major role in embryo development, but there is a limited understanding of the connection between sperm quality, embryo development, and morphokinetic parameters using donor oocytes. The study evaluated if sperm quality may influence the morphokinetic parameters in IVF cycles. A retrospective multicentric observational cohort study included 747 ICSI cycles using donor oocytes with fresh or frozen sperm. Embryos were cultured in time-lapse incubators until the blastocyst stage. The population was divided into three groups according to sperm concentration, as control group (> 16 mill/mL), severe oligospermia (0-5 mill/mL), and moderate oligospermia group (5-16 mill/mL). Morphokinetic analysis showed no difference in the time from the 2-cell to 6-cell stage of embryo development. A significant difference was observed on day 3 of embryo development, specifically at the 7-cell stage (t7), severe oligospermia 53.37 ± 9.81, moderate oligospermia 56.95 ± 9.78, and control 55.1 ± 8.85h post-insemination (hpi) (p = 0.024), and 8-cell stage (t8), severe oligospermia 55.41 ± 10.83, moderate oligospermia 61.86 ± 12.38 hpi (p < 0.001), and control 58.61 ± 11.33. Accordingly, the synchrony of the four cleavages going from 4 to 8 cells (s3) was found statistically different among the groups in the severe oligospermia 8.05 ± 9.99, moderate oligospermia 11.66 ± 11.04 hpi, and control 8.55 ± 8.58 (p = 0.009). Morphokinetic time ranges were obtained for t6, t7, t8, and s3 in order to identify the good-quality blastocysts. Poor sperm quality is associated with alterations in morphokinetic parameters on day 3 in IVF cycles with donor oocytes, underlining the important role of spermatozoa during embryo development.

Is there a relationship between morphokinetic parameters and obstetrical complications? An analysis of singleton live births after single fresh embryo transfer

BackgroundWith the advancement in embryology and the introduction of time-lapse monitoring system, the embryologists' goal might be to find not only the embryo with the highest probability of live birth, but also the embryo with the highest probability to progress to a healthy full-term delivery. Thus, we aimed to investigate the association between morphokinetic time-lapse parameters and obstetrical and perinatal complications.MethodsA cohort study reviewing fertility and delivery files of all singletone births from IVF patients whose embryos were cultured in a time-lapse monitoring system and had a single fresh embryo transfer at our center between 2013–2019. We conducted multiple comparisons between complicated and uncomplicated pregnancies of each perinatal complication, including: gestational diabetes mellitus (GDM); small for gestational age (SGA); pre-eclamptic toxemia (PET); preterm labor < 37 weeks of gestation (PTL); and third stage of labor complications. A comparison between pregnancies with and without a composite outcome of placental complications including GDM, SGA, PET and PTL was also conducted. Baseline characteristics, treatment and morphokinetic parameters in complicated and uncomplicated gestations were compared. Logistic regression analysis was utilized to adjust results for potential confounders.ResultsOne hundred seventy-six single embryo transfers resulted in 176 live births. Morphokinetic time-lapse parameters were similar between the groups, except for a shorter time to full blastulation in the SGA group (tB-tPNf = 75.5 ± 1.3 h vs. 79.5 ± 4.8 in the non-SGA group, p < 0.001), and shorter third cell cycle duration in the PET group (CC3 = 12.4 ± 1.1 h vs. 13.6 ± 2.9 in the non-PET group, p = 0.02). On multivariate regression analysis, none of the morphokinetic parameters were found to be significantly correlated with any of the perinatal complications.ConclusionTime-lapse morphokinetic parameters of the embryo transferred are not associated with adverse obstetric and perinatal outcomes.

Expression of exosomal microRNAs miR-34a and miR-210 in male infertility: relationship with morphokinetic parameters and sperm DNA fragmentation

Introduction. Infertility affects tens of millions of men and women across the globe. In approximately half of cases, male factors are the cause of infertility. In recent decades, there has been a significant decline in the quality of male ejaculate, which is characterized by reduced sperm concentration and motility. The insufficient diagnostic and prognostic value of routine semen analysis results highlights the challenge of developing effective diagnostic tools and searching for reliable biomarkers of male infertility. One of the most promising approaches may be assessing sperm microRNA expression.Objective. To study the role of sperm exosomal microRNAs miR-34a and miR-210 in the development of male infertility.Materials &amp; methods. The retrospective study included 150 men aged 25 – 49 years; of these, 96 patients were diagnosed with idiopathic infertility. The comparison group consisted of 54 fertile men. To assess the structure and motility of sperm, the results of a standard ejaculate study (WHO, 2021) and computer data analysis using MMC Sperm (MMCSoft, St. Petersburg, Russia) software were used. The degree of DNA fragmentation was assessed using the TUNEL method. To analyze the expression of miR-34a and miR-210, quantitative real-time PCR was performed using the miRCURY LNA SYBR Green PCR Kit («Qiagen» GmbH, Hilden, Germany) and the Rotor-Gene Q PCR product detection system («Qiagen» GmbH, Hilden, Germany).Results. The study of ejaculate using the Computer Assisted Sperm Analysis System (CASA) method revealed a statistically significant relationship between the level of DNA fragmentation of sperm and indicators of the speed of their movement: rectilinear (VSL) (r = -0.522726; p &lt; 0.01), curvilinear (VCL) (r = -0.499096; p &lt; 0.01), along the middle path (VAP) (r = -0.429533; p &lt; 0.01), as well as with the amplitude of lateral head displacement (ALH) (r = -0.294779, p &lt; 0.01), the linearity of their curvilinear path (LIN) (r = -0.385796; p &lt; 0.01), the degree of straight-line movements (STR) (r = -0.268248; p &lt; 0.05) and their progressive mobility (r = -0.411547; p &lt; 0.01). A study of the level of microRNA expression in sperm exosomes revealed a statistically significant decrease in its miRNA-34a pool (p = 0.0116). According to the Chaddock scale, the strength of the correlation between miR-210 expression and the effectiveness of ART programs was moderate (0.437993). The inverse relationship between miR-34a expression and IVF and ICSI results was weak (0.135314).Conclusion. The analysis of exosomal microRNA-34a and microRNA-210, which are involved in the regulation of spermatogenesis, reveals a direct correlation between their variations and changes in the kinetic and morphological parameters of gametes. It also indicates a relationship with the state of DNA fragmentation. These findings suggest varying levels of gene expression among infertile patients, men with proven fertility, and those undergoing assisted reproductive technology (ART) treatments, both with successful and repeated unsuccessful outcomes.

The synergy of morphokinetic parameters and sHLA-G in cleavage embryo enhancing implantation rates.

Objective: This study aimed to assess the relationship between implantation and soluble HLA-G (sHLA-G) expression in cleavage embryo culture medium (ECM) in conjunction with early developmental kinetics determined by time-lapse imaging (TLI). Methods: A retrospective, single-center study was conducted involving 238 embryos from 165 patients who underwent Frozen-thawed embryo transfer (FET) using autologous oocytes, with either single or double embryo transfer. TLI morphokinetic parameters (t2, t3, t4, t5, t6, t7, t8, cc2, s2, cc3, s3) of embryos were analyzed, and sHLA-G levels in D3 ECM were measured using an enzyme-linked immunosorbent assay (ELISA). A hierarchical classification model was developed to categorize embryos into five groups (A, B, C, D, E). The correlation between sHLA-G levels, TLI classification of embryos, and embryo implantation was investigated to establish a non-invasive method for evaluating implantation potential. Multivariate logistic regression analysis was performed to identify potential influencing factors, and receiver operating characteristic (ROC) curves were used to evaluate the predictive value for implantation. Results: Multivariate unconditional logistic regression analysis indicated that TLI parameters t5 and s3 and sHLA-G level in ECM were independent risk factors affecting embryo implantation. The implantation rate decreased from TLI classification A to E. The proposed classification model effectively assessed the implantation potential of embryos. The implantation rate was higher in the sHLA-G positive group compared to the sHLA-G negative group (p < 0.001). The expression of sHLA-G in D3 ECM, combined with the TLI classification model, accurately evaluated the implantation potential of embryos with an AUC of 0.876. Conclusion: The integration of cleavage kinetics and embryonic sHLA-G expression could reliably identify embryos with a high likelihood of successful implantation.

P-615 Can the use of progestin in egg donation cycles influence morphokinetic parameters of embryos when compared with GnRH antagonist during ovarian stimulation? A case-control study

Abstract Study question Are there differences in clinical and morphokinetic parameters of embryos obtained from fresh egg donation cycles using GnRH antagonist versus PPOS on same donor? Summary answer Lower number of oocytes retrieved and faster early morphokinetics parameters were observed in PPOS protocol. What is known already Controlled ovarian hyperstimulation (COH) is a crucial step in assisted reproduction treatments and the ideal protocol should be chosen according to the patient’s profile, based on literature results, intending to achieve high effectiveness. The use of PPOS in ovarian stimulation protocols is getting attention because of its safety, efficacy and proved to be patient friendly and an economical alternative to avoid premature LH surge. In the last years, morphokinetics parameters started to be considered in embryo classification and selection, as they correlate to embryo implantation potential. Here, morphokinetics parameters were compared from two different COH in donated oocyte cycles. Study design, size, duration Case-control study including cycles from 10 oocyte donors and 20 recipient patients, whose embryos formed with fresh oocytes were cultured in a time-lapse system (Embryoscope® Plus), in a single private IVF center between Jun/2020 and Oct/2022. Each donor were submitted to two cycles: first stimulation with standard GnRH antagonist (ant) short protocol (group A) and in a posterior cycle: a second stimulation with progesterone (PPOS protocol with dydrogesterone, group B). Participants/materials, setting, methods All donors used menotropin for COH (doses adjusted according to patient’s profile). Group A started GnRh ant with follicle above 14 mm. On group B, dydrogesterone 10 mg was started on the first day of COH and maintained until the day of ovulation trigger, which was performed with GnRH agonist. Clinical and morphokinetics parameters were compared. Paired t-test, t-test or Mann-Whitney were used properly, values of p &amp;lt; 0.05 were considered significant. Main results and the role of chance A total of 70 blastocysts were analyzed, 39 from group A and 31 from group B. The following morphokinetics parameters were annotated: time of PN fading (tPNf), time to two cells (t2), three cells (t3), four cells (t4), five cells (t5), eight cells (t8), time to morula (tM), time to blastulation (tB) and KIDScoreTM. Statistically significant differences were observed for tPNf: group a) 24,01 ± 3,28h versus b) 22,59 ± 2,48h, p = 0,0484, t2: a) 26,30 ± 3,24h versus b) 24,76 ± 2,27h, p = 0,0211; t3: a) 37,04 ± 4,44h versus b) 35,53 ± 2,93, p = 0,0347 and t4: a) 38,38 ± 5,15h versus b) 36,05 ± 3,32h, p = 0,0152. Clinical characteristics of the donors were also analyzed: age, body mass index (BMI), total gonadotropin dose, antral follicular count (AFC), follicle stimulating hormone (FSH), basal hormone levels, follicle count, oocyte pick-up (OPU), altered oocytes, number of germinal vesicles, metaphase I (MI) and metaphase II (MII) oocytes. Differences were observed only between two clinical aspects: patients’ age: a) 24,2 ± 3,6 versus b) 25,0 ± 3,3, p = 0,0031 and OPU: a) 23,3 ± 9,1 versus b) 18,6 ± 8,3, p = 0,004, however both variables does not seem to affect embryo formation potential. Limitations, reasons for caution The results should be interpreted with caution due to the limitations of evaluate a small dataset of donors, resulting in low number of embryos in each group of protocols for analysis. Wider implications of the findings In this case-control study, a lower number of oocytes retrieved and faster early morphokinetics times were observed in PPOS compared to antagonist protocol. The implications of these differences still need further data analyses to compare if these findings can change clinical outcomes. Trial registration number Not applicable.

P-186 Embryos developing within the optimal morphokinetic range have a 5% higher chance of leading to a Live birth (LB)

Abstract Study question Is there an optimal range of morphokinetic parameters that have a higher LB rate, if so what is the percentage of LB improvement? Summary answer Optimal embryo development range in morphokinetic parameters was identified in t6, t8, tB, t8-t4 and t8-t2 leading to a 5% improvement in LB. What is known already Achieving LB is the main objective in IVF. Traditional embryo assessment based on morphology has drawbacks, with variability among embryologists and limited predictive ability for a single-day assessment. The emergence of AI in IVF, exemplified by CHLOE-EQ, an embryologist’s tool to register embryo morphokinetics and other parameters, presents a solution. This study investigates whether AI-detected morphokinetic parameters can significantly improve LB rates, providing a more precise and predictive approach to embryo assessment. This study investigates if embryos that lead to LB develop under specific morphokinetic optimal ranges. Study design, size, duration Conducted across multiple centres within a private clinic, this retrospective study investigated 7156 embryos with known birth outcome between January 2021 and January 2023. Participants/materials, setting, methods Absolute (tPNf to tEB) and interim morphokinetic parameters that have been previously been found to impact embryo development (CC1, CC2, CC3, S2, S3, t5-t2, t8-t4, t8-t2, tEB-tSB) were automatically annotated by an AI model (CHLOE-EQ, Fairtility Ltd). The interquartile range (Q1-Q3) of embryos that resulted in live birth was analyzed to identify the optimal range. Chi-square analysis compared live birth rates between embryos within and outside this range. Main results and the role of chance A significant Live birth optimal range was found for t6 (47.8-57.3), t8 (51.8-65.8), tB (100.9-113.6), t8-t4 (15.2-25.5), t8-t2 (26.9-39.2), p &amp;lt; 0.05. The following morphokinetic parameters were associated with higher LB rate when the embryo was within the optimal range hours as compared to outside the optimal range: t6 (38%, n = 1012 vs 35% n = 1007, p &amp;lt; 0.05), t8 (38%, n = 947 vs 35%, n = 941, p &amp;lt; 0.05), tB (33%, n = 548 vs 30%, n = 533, p &amp;lt; 0.05), t8-t4 (38%, n = 926 vs 34%, n = 943, p &amp;lt; 0.05), t8-t2 (38%, n = 938 vs 35% n = 936, p &amp;lt; 0.05). No significant LB rate differences were observed for other morphokinetics. Overall, LB rates improved by 3-5% in embryos developing at the optimal rate, with a slightly higher significance in absolute morphokinetic values over interim ones. Limitations, reasons for caution The study’s retrospective nature entails inherent biases. Additionally, this analysis does not account for oocyte origin and male factor as bias elements, which will be considered in future research. Thus results may not be transferrable to clinics that have other clinical practices and different patient demographics. Wider implications of the findings Certain morphokinetic parameters can be added to the already existing range of available LB evaluation methods. Future studies should explore diverse morphokinetic ranges to identify additional optimal intervals for each parameter. AI tools can facilitate automatic annotation and measurement of embryo morphokinetics, enhancing precision in assessment. Trial registration number Not applicable

O-157 The effect of growth hormone on the euploid rate and morphokinetic signature of blastocysts in patients with advanced maternal age : a randomized controlled trial

Abstract Study question Does growth hormone (GH) supplementation during ovarian stimulation affect the euploid rate and morphokinetic characteristics of euploid embryos in patients with advanced maternal age (AMA)? Summary answer GH supplementation improved the probability of obtaining euploid blastocysts in women with AMA, while did not affect the morphokinetic parameters of the euploid embryos. What is known already Studies have demonstrated that supplementation with GH during ovarian stimulation (OS) has positive impact on the clinical outcomes of in vitro fertilization (IVF). However, few studies have investigated the association of GH supplementation with the ploidy status of embryos. Meanwhile, it is unclear whether GH supplementation affects the morphokinetic signatures and the transfer outcomes of euploid blastocysts. Therefore, the underlying reason for the effect of GH on IVF outcomes remains largely unknown. Study design, size, duration This is a single centre open-labelled randomized controlled trial conducted at single IVF center from August 2022 to August 2023. A total of 441 patients were screened for eligibility and 309 patients were enrolled and allocated. The primary outcome was the proportion of cycles which obtained euploid blastocysts. Secondary outcomes included euploid blastocyst rate per cohort, euploid blastocyst rate per cycle and morphokinetic parameters of euploid embryos obtained with or without GH supplementation. Participants/materials, setting, methods A total of 309 patients aged 38–46 years who underwent their first preimplantation genetic testing for aneuploidy (PGT-A) cycle using antagonist protocol were randomized into the GH group or the control group at 1:1 ratio, with all embryos cultured in the time-lapse monitoring system. Patients in the GH group received 2 IU/day subcutaneous GH from the 2nd day of the previous menstrual period until the day of trigger, for a total of approximately 6 weeks. Main results and the role of chance The baseline and cycle characteristics between the two groups were similar. The proportion of cycles which obtained euploid blastocysts was significantly higher in the GH group (44.74% vs 31.85%, p = 0.02). The GH group had a significantly higher euploid blastocyst rate per cohort (38.46% vs 28.01%, p = 0.01) and mean euploid blastocyst rate per cycle (per biopsy cycle 0.38±0.43 vs. 0.26±0.39, p = 0.02; per OS cycle 0.29±0.41 vs. 0.19±0.36, p = 0.03). Multivariate logistic regression indicated GH supplement was an independent factor for acquisition of euploid embryos (adjusted risk ratio=1.79, 95%CI [1.06-2.99], p = 0.03). We did not observe a statistically significant difference in the morphokinetic parameters and transfer outcomes of the euploid embryos acquired with or without GH supplemented. Limitations, reasons for caution This was a single center study which limited the sample size, further limiting the number of patients with euploid blastocysts analyzed for the morphokinetic parameters and transfer outcomes. Meanwhile, affordability may also introduce selection bias since patients were required to pay for GH. Wider implications of the findings Our research presents new evidence that GH supplementation AMA patients increased the probability of obtaining euploid blastocysts, and explains the underlying reason for the positive effect of GH on IVF outcomes in existing studies. The effect of GH on morphokinetic parameters and transfer outcomes of euploid embryos warrants further investigation. Trial registration number NCT05447208, www.clinicaltrials.gov.

P-222 Can embryo morphokinetics act as early warning key performance indicators in relation to consumable batching

Abstract Study question Can morphokinetics be used as an early warning indicator of a batch-related effect of oil currently in use in the laboratory on implantation rate? Summary answer Where morphokinetic timings were slower for a batch of oil, implantation rate was also reduced, suggesting they could be used as an early warning system. What is known already The commercialisation of oil means that there are strict manufacturing guidelines leading to consistency between batches. However, this does not mean that minor differences do not exist. Furthermore, oil can undergo peroxidation if stored incorrectly, causing it to become toxic to embryos. The introduction of timelapse incubation has allowed extensive research into several morphokinetic parameters. This research has established time-frames within which each developmental milestone should occur to indicate the likelihood of implantation. Implantation rate (IR) is often used as a key performance indicator in the laboratory however, potential systemic issues may be identified sooner if morphokinetic parameters were used. Study design, size, duration The following morphokinetic parameters were compared to IR to determine if a batch of oil was affecting embryo development; time of division to two cells (t2), five cells (t5), eight cells (t8), time to start of blastulation (tSB) and cell synchronicity between two and eight cells (CS2-8). Data from January 2019 to October 2019 from a single centre were used, and included data from 2008 embryos and 333 patients. Participants/materials, setting, methods Data was exported from the EmbryoScope + [Vitrolife, Sweden]. Data from RIWitness Laboratory Manager [CooperSurgical, Denmark] was utilised to identify when different batches of oil (Ovoil™, Vitrolife) were used within the specified time period. The data from the EmbryoScope+ was then separated by date to signify when a new batch of oil was open. A Kruskal-Wallis test was then used to identify whether the morphokinetic timings differed between batches of oil. Main results and the role of chance There was no significant difference in patient age (p = 0.178) or the number of oocytes collected (p = 0.097) between batches of oil used. There was no significant difference between the morphokinetic timings for each batch of oil used. However, the morphokinetic timings follow the same trend line as the IR for both t2 and t8; where t2 and t8 are higher, the IR for the corresponding batch of oil is reduced. The trend can also be seen in CS2-8. The batch of oil with the lowest IR (25.56%) also had the highest mean time to t2, t8 and tSB compared to the other batches of oil. This corresponds with previous research that identified a correlation between slower embryo development and reduced IR (Meseguer et al., 2011). A clear difference could be seen between the average time taken to reach each developmental milestone and each batch of oil used, perhaps suggesting oil can have an effect, however small, on embryo development. This study suggests that morphokinetic timings could be used as an early warning indicator that a new batch of oil may be affecting development allowing a rapid response by the laboratory team and possible cessation of use pending further investigation. Limitations, reasons for caution Different batches of culture medium were also used throughout this time period and were not introduced at the same time as oil batches. Therefore, the results of this investigation cannot be associated entirely to the introduction of a different batch of oil. Wider implications of the findings This information could be incorporated into an early warning algorithm that could be assessed after a batch of oil was introduced. The use of the batch could then be ceased immediately without waiting for a decrease in IR by which time it would have been used for many more embryos. Trial registration number Not applicable

P-163 Does severe endometriosis (SE) have an effect on embryo morphokinetic parameters?

Abstract Study question Does severe endometriosis (SE) have an effect on embryo morphokinetic parameters? Summary answer Severe endometriosis may have an effect on embryonic cell divisions and could change morphokinetic parameters of in vitro developing embryos. What is known already Endometriosis is one of the leading causes of infertility that affects approximately one third of women who apply for in vitro fertilization (IVF) treatment. Controversial results were reported regarding the effect of endometriosis on both morphological and dynamic characteristics of embryos so far. Study design, size, duration This was a retrospective comparative analysis including a total of 1280 embryos (SE: 729, Tubal factor (control): 551) of 241 patients (SE: 134, control: 107) that were incubated at Time-Lapse Monitoring System (TLM) in Memorial Sisli Hospital, ART and Reproductive Genetics Center between October 2011 and July 2023. Patients with ≥ 38 years, BMI ≥ 30 and partners having severe male factor (sperm concentration: &amp;lt;5 mil/ml) were excluded from the study. Participants/materials, setting, methods Severe endometriosis was defined as Grade III and IV according to ASRM classification. IMSI (Intracytoplasmic Morphologically Selected Sperm Injection) was performed to all oocytes. Embryos were evaluated according to Gardner’s classification. Fertilization rates were assessed using Chi-square test. Morphokinetic parameters of embryos were compared by Student’s t-test using SPSS 28.0. Main results and the role of chance There was a homogenous distribution in terms of female age, BMI, basal FSH, AMH levels, total and mature (MII) oocyte count between study groups. Abnormal fertilization rate was higher in SE group than controls ( 4.3% vs 2.4%, p = 0.035). All morphokinetic parameters were found significantly delayed in the SE group than in the control group (p &amp;lt; 0.05), however, no statistical significance was obtained in terms of tSC and tEB (p &amp;gt; 0.05). Duration of first, second and third embryo cell cycles (ECC1, ECC2, and ECC3), synchronization of cell divisions (S2: t4-t3) and cleavage patterns (S3: t8-t5) were also found statistically significant between SE and control groups (p &amp;lt; 0.05). Limitations, reasons for caution The major limitation of this small-sized, retrospective study is that clinical outcomes of the embryos were not assessed. Wider implications of the findings Delayed development was observed in the embryos of SE patients. Prospective studies with larger cohorts are needed to have a better understanding between severe endometriosis and embryo morphokinetics. Trial registration number Not applicable

O-248 Novel early cleavage parameters improve embryo selection and reach ploidy predictive values comparable to artificial intelligence

Abstract Study question Can we improve embryo selection and ploidy prediction using novel morphokinetic parameters in blastocyst development, and obtain results comparable to artificial intelligence? Summary answer Time and pattern of novel parameter st2 are related to ploidy. Including them in embryo evaluation improves ploidy prediction, reaching predictive values comparable to AI. What is known already Time-lapse has allowed us to detect more morphokinetic parameters and improve embryo grading systems. However, ploidy prediction still relies on invasive techniques to obtain a reliable result. Artificial intelligence is promising and can already predict blastulation or implantation with good results, but not ploidy yet. Moreover, AI presents some other limitations now, including the black box behind its decisions, which doesn’t allow clinicians to know what parameters are being considered the most. Improving the manual evaluation tools used by embryologists can help improve clinical outcomes until AI is fully introduced in IVF. Study design, size, duration A single center retrospective study was performed. Data included a total of 608 blastocyst from 106 treatments between January 2018 and July 2022. Participants/materials, setting, methods 160 treatments of patients for preimplantation genetic testing cycles were studied. Embryos were cultured in time-lapse incubators and were studied for both morphologic and kinetic parameters. 608 of the embryos developed to good quality blastocyst and were able to be biopsied, and therefore analyzed by PGT for aneuploidies on Day5 o Day6. Annotation of several morphokinetic characteristics was performed on all analyzed embryos. The study was blinded for ploidy and all clinical outcomes. Main results and the role of chance Improved embryo grading systems used by embryologist can reach high predictive values comparable to artificial intelligence. We present a new parameter, “st2, start of t2”, which refers to the first cytoplasmic movements detected at the beginning of the first cell cleavage, as highly implicated in ploidy status. Earlier st2 is associated to better euploidy rates (p &amp;lt; 0.001). Not only the time, but also the phenotype of the movements appears as related to ploidy. Embryos showing none or un-patterned random cytoplasmic movements show a marked tendency towards aneuploidy. Contrary, circular waves prior to cell division are highly associated with euploidy (90% of the cases) (p &amp;lt; 0.0001). The implication of the described parameters was also confirmed by logistic regression, with a receiver operating curve (ROC) value of 0.69 for ploidy prediction (95% confidence interval (CI), 0.62 to 0.76). We processed raw time-lapse images with an automatic annotation software using artificial intelligence and obtained a ploidy predictive value of 0.64. This means that using the optimal parameters in manual embryo evaluation, we are able to improve our grading systems and reach ploidy predictive values comparable than with AI. Limitations, reasons for caution This is a retrospective study. The ploidy predictive potential is now being tested on a prospective study. Wider implications of the findings Optimizing the indicators to select the most suitable blastocyst, like including st2, can help reducing the time until the pregnancy of an euploid baby, avoiding invasive and expensive methods. Moreover, by combining AI with the novel parameter st2 we could improve the potential of the ploidy predictions. Trial registration number not applicable

P-515 The morphokinetic signature of high-level mosaic embryos in time-lapse ART analysis

Abstract Study question Do high-level mosaic embryos exhibit differences in developmental speed compared to euploid or low-level mosaic embryos? Summary answer High-level mosaic embryos exhibit slower development than euploid and low-level mosaic embryos during the blastulation and blastocyst formation stages. What is known already Time-lapse technology (TLT) has facilitated the real-time observation of human embryo development without disrupting culture conditions, generating a wealth of morphokinetic data linked to outcomes such as implantation potential, blastocyst development, and ploidy status. However, the developmental patterns of mosaic embryos based on their aneuploidy level using TLT remain not well understood. Consequently, we aimed to assess the clinical significance of morphokinetic parameters in human embryo development, with a specific focus on embryos categorized by their mosaic status. Study design, size, duration This retrospective study examined 385 blastocysts undergoing preimplantation genetic testing for aneuploidy (PGT-A) between January and December 2022. Out of the total, 191 expansion blastocysts underwent PGT-A biopsy on day 5. The PGT-A results, which categorized embryos into euploid, aneuploid, and mosaic based on their genetic composition, were determined through next-generation sequencing (NGS) by an analysis company. Mosaic embryos were further categorized into two groups: high-level mosaicism (&amp;gt;30% aneuploidy cells) and low-level mosaicism (≤30%). Participants/materials, setting, methods Differences in development rates were assessed through the examination of morphological and morphokinetic parameters. These parameters included the time to second polar body extrusion (tPB2), time to pronuclei appearance and fading (tPNa, tPNf), time to reach 2–8 cells (t2, t3, t4, t5, t6, t7, t8, t9), and time to the onset of blastulation and blastocyst formation (tSB, tB). Statistical analyses, incorporating ANOVA and post-hoc tests, were employed to compare developmental timings across different groups. Main results and the role of chance When categorizing mosaic embryos based on their level, high-level mosaic embryos showed a significantly slower developmental rate than low-level mosaic embryos, evident at both the blastulation stage (tsB, high-level vs. low-level: 101.80h vs. 97.3h, p=0.003) and blastocyst stage (tB, high-level vs. low-level: 109.98h vs. 105.92h, p=0.005). In comparison to euploid embryos, low-level mosaic embryos demonstrated comparable outcomes. Conversely, high-level embryos exhibited a significant difference in both blastulation (tsB, euploidy vs. high-level: 98.85h vs. 101.80h, p=0.026) and blastocyst (tB, euploidy vs. high-level: 106.73h vs. 109.98h, p=0.011) stages. These findings suggested that the mosaic level of embryos might influence developmental kinetics, proposing this approach as an additional tool for the selection of mosaic embryos in the context of embryo transfer. Limitations, reasons for caution This study focused on morphokinetic parameters captured using time-lapse imaging, which may not fully represent the complex cellular dynamics within embryos. Furthermore, the categorization of mosaic embryos into high-level and low-level groups based on the percentage of mosaicism (&amp;gt;30% and ≤30%, respectively) might not capture all embryo developmental potential. Wider implications of the findings The findings suggest valuable predictive potential in assessing the development of aneuploid and mosaic embryos. Further research into the development of euploid, aneuploid, and mosaic embryos, as well as high-level and low-level mosaic embryos, could provide valuable insights for ART practices. This could significantly impact ART decision-making and patient outcomes. Trial registration number not applicable

P-249 Morphokinetic parameters comparison between balanced and unbalanced embryos from patients with structural chromosome abnormalities

Abstract Study question Are there morphokinetic differences between balanced and unbalanced embryos after PGT-SR? Summary answer The total time embryos remain at the morula stage (TiCAV- TM) could be a morphokinetic marker to consider in PGT-SR cycles. What is known already Due to the preimplantation genetic testing for structural rearrangements (PGT-SR) it is possible to identify embryos with an excess or defect of genetic material associated with a parental structural rearrangement. This technique is suitable for patients with a normal phenotype but a balanced chromosomal rearrangement. These patients have a higher risk of spontaneous abortion, as well as higher aneuploidy rates. Within structural chromosomal alterations, translocations are the most common presentation of chromosomal rearrangement. Study design, size, duration Within this multicentre retrospective study, we evaluated 133 embryos, from 39 cycles in which one of the partners had a karyotype with balanced reciprocal or Robertsonian translocations, from January 2021 to September 2023. After PGT-SR was carried out, we obtained 40 balanced embryos and 93 unbalanced embryos. Participants/materials, setting, methods We compared the conventional morphokinetic parameters and the duration of each state between two groups after PGT-SR samples were analysed. All embryos were cultured in Geri® time-lapse incubators, and underwent embryo biopsy on day 5-7. Statistical analysis was conducted using R software (v.4.3.1), including the following confounding variables: origin of karyotype alteration (maternal or paternal), type of translocation (Robertsonian or reciprocal), female and male gamete age, type of oocyte (fresh or frozen) and oocyte origin. Main results and the role of chance We observed that the mean age of the oocyte was significantly higher in the unbalanced embryos compared to the balanced ones (32.85±5.32 years vs.30.45±5.01 years). In terms of the altered karyotype, the paternal origin is more prevalent than the maternal one (60.9% vs. 39.1%). Regarding the type of translocation, reciprocal translocation is more predominant than Robertsonian translocation (83.5% vs. 16.5%) in our study population. When comparing morphokinetic parameters between both groups, we observed that unbalanced embryos tend to reach an expanded blastocyst stage later than balanced embryos (107.62±10.02h vs. 102.48±8.87h, respectively; p = 0.008, corrected p-value=0.186). We observed the same tendency with hatching initiation time (110.64±10.49h vs. 105.91±9.38h; p = 0.013, corrected p-value=0.296). We observed that unbalanced embryos have a slower development than balanced embryos (TiCAV-TFB 10.10±4.46h vs. 8.35±3.79h; p = 0.006, corrected p-value=0.437; TiH-TiCAV 13.03±5.98h vs. 11.79±5.50h; p = 0.017, corrected p-value=0.816). We did observe significant differences in total time at morula stage (TiCAV – TM) when comparing unbalanced and balanced embryos (12.60±6.94 vs. 8.88±5.65; p = 0.003, corrected p-value=0.001). Limitations, reasons for caution A higher number of embryos should be evaluated in order to enhance the statistical power of this study, as there is an important difference in sample size between balanced and unbalanced embryos in the PGT-SR cycles included in this study. Wider implications of the findings Further morphokinetic study of embryos from patients with structural chromosomal abnormalities can open up new lines of research and can help us to predict the final outcome of the cycle, accompanied by a genetic study. Trial registration number Not applicable

P-139 Mitochondrial DNA copy number in trophectodermal cells (MITOSCORE) is related to embryo quality, but not to implantation potential

Abstract Study question Is mtDNA copy number in trophectodermal cells (MITOSCORE) related to the quality of euploid embryos in our PGT-A program? Is it related to implantation potential? Summary answer mtDNA copy number in trophectodermal cells (MITOSCORE) is related to embryo quality and morphokinetic, but no to implantation potential in our PGT-A program. What is known already Fertilization and embryonic development require energy, sourced from mitochondria. Research on human embryos indicates associations between aneuploidy and maternal age and increased mtDNA content. Consequently, exploring embryonic mtDNA levels as a quality marker gained momentum. Embryos with high mtDNA levels often correlate with poor implantation potential, leading to its adoption as a biomarker for selection. However, conflicting studies challenge this notion, failing to establish correlations between mtDNA levels and blastocyst grade, implantation, or ongoing pregnancy rates in patients with euploid embryos. Consequently, the predictive value of mtDNA levels for embryo implantation potential and pregnancy outcomes remains uncertain. Study design, size, duration A retrospective study was conducted to establish a relationship between MITOSCORE and the euploid embryo morphokinetics. It involved 59 embryos from 25 PGT-A cycles performed from January 2022 to December 2023. In addition, we retrospectively analysed 143 thawed single embryo transfers from three centres (October 2020 to December 2023) to assess the correlation between MITOSCORE and pregnancy outcomes. Embryos for transfer were selected based on morphology; mtDNA was considered only if two had similar morphology. Participants/materials, setting, methods Morphokinetic was determined using the AI tool CHLOE (Fairtility, Israel), which automatically provides morphokinetic parameters and also offers an additional score of embryo quality, EQ. mtDNA levels were estimated by NSG and the MITOSCORE score was provided by an algorithm (Igenomix, Spain). Pearson’s correlation analysis was employed to investigate the potential relationship between MITOSCORE results and embryonic morphokinetics. Mann-Whitney test was used to analyse the relationship of MITOSCORE levels with pregnancy. Main results and the role of chance A statistically significant correlation (p &amp;lt; 0.05), albeit weak, was observed between MITOSCORE and embryonic quality evaluated through CHLOE EQ (r: -0.3858). It is noted that a higher EQ was associated with a lower value in MITOSCORE. Additionally, upon analysing morphokinetic parameters, it was found that the time to reach the stages of morula (r: 0.6876), early blastocyst (r: 0.6303), cavitated (r: 0.6823), and expanded (r: 0.5776) were significantly correlated (p &amp;lt; 0.05) with MITOSCORE values. The slower the embryo development, the higher the MITOSCORE. Regarding implantation capacity and mtDNA levels, implanted euploid embryos had a MITOSCORE value of 20.35, which was not significantly different from that of non-implanted embryos, which was 21.65. The highest MITOSCORE value at which a full-term pregnancy was achieved was 32.34. However, no pregnancy was obtained with the three embryos with the highest scores (35.02, 45.02, and 48.17). Limitations, reasons for caution Morphokinetics correlation with MITOSCORE was analysed with a limited number of embryos (59). The highest MITOSCORE level was 48.17. The transfers were not fully blinded. A randomized clinical trial would be necessary to determine the true clinical benefits of mtDNA levels in the trophectoderm of euploid embryos in our clinics. Wider implications of the findings Trophectodermal mtDNA level is associated with morphokinetic parameters and blastocyst quality but not embryo implantation potential. In our PGT-A programme, and with our results, trophectodermal mtDNA copy numbers work as a biomarker of blastocyst quality but have not been correlated with clinical outcomes after the transfer of euploid embryos. Trial registration number NOT APPLICABLE

P-264 Analysis of embryo development based on time of pronuclei appearance (tPNa) and aspects of overcoming delayed pronuclei appearance through morphokinetic patterns

Abstract Study question How much delay in the appearance of 2PN can be considered normal when predicting embryo transfer potential? Summary answer Embryos with tPNa &amp;lt;13hpi post-insemination develop into good-quality blastocysts; embryos with tPNa ≥13hpi develop blastocysts by shortening the time of tPNf-tPNa. What is known already Oocytes generally exhibit 2PNs at 17–20 hours, with pronuclei fading (time of pronuclei fading; tPNf) at 23–25 hours and developing into two cells (time of 2-cell division; t2) at 25–33 hours. Fertilization confirmation is typically performed 17–18 hours after insemination. However, a few oocytes with no visible PN (0PN) at the time of fertilization confirmation develop morphologically normal blastocysts, ultimately leading to pregnancy. The failure to identify PNs can be attributed to two scenarios: rapid fading or delayed appearance. Notably, there is still a lack of research on the normal range for delayed PN appearance. Study design, size, duration This study was conducted with 2153 embryos obtained from 390 intracytoplasmic sperm injection cycles (August 2021 to June 2023). All embryos were incubated for 5 days using a time-lapse system (EmbryoScopeTM+, Vitrolife). The blastocyst development rate and morphokinetic parameters according to the tPNa of embryos were analyzed using KIDScore D5 v3 and iDAScore v2.0 (VTH server+, Vitrolife). Clinical pregnancy was also analyzed. Participants/materials, setting, methods Morphokinetic parameters were analyzed from time of second polar body (tPB2) ∼ time of expanded blastocyst (tEB). The times taken for PN to appear after the second polar body release (tPNa–tPB2), for PN to fade (tPNf–tPNa) and for 2-cell division (t2–tPNf) were calculated and compared. Blastocysts were graded using the Gardner system, a grade of BB or higher divided into good quality blastocysts (GQ-BL). Main results and the role of chance tPNa was observed as 8.09±2.11hpi [1.93hpi∼32.46hpi; &amp;lt;5hpi (n = 49), 5∼6hpi (n = 474), 7∼8hpi (n = 1220), 9∼10hpi (n = 266), 11∼12hpi (n = 91), 13∼14hpi (n = 29), 15∼20hpi (n = 18) and &amp;gt;20hpi (n = 6)]. The rate of blastocysts was highest at 5∼6hpi (64.98%) and significantly lower at 9∼10hpi (54.14%), 11∼12hpi (42.86%) and 13∼14hpi (31.03%) (p &amp;lt; 0.005). Similarly, the rate of GQ-BL was also highest at 5∼6hpi (29.11%) and significantly lower at 9∼10hpi (19.17%), 11∼12hpi (10.99%) and 13∼14hpi (3.45%) (p &amp;lt; 0.005). No embryos developed into GQ-BL at 15∼20hpi, and no embryos developed into blastocysts at &amp;gt; 20hpi. The iDAScore was significantly different at &amp;lt; 13hpi and ≥13hpi (6.00±1.86 vs. 4.24±2.26, p &amp;lt; 0.005). Similarly, KIDScore D5 showed the same patterns (6.33±1.93 vs. 4.05±1.66, p &amp;lt; 0.005). At ≥ 13hpi, no blastocysts led to pregnancy. Morphokinetic parameters were analyzed to identify the factors influencing the development of blastocysts. The analysis revealed that tPNf-tPNa tended to gradually become shorter with delayed tPNa. Notably, there were significant differences in tPNf-tPNa between &amp;lt;13hpi and ≥13hpi (16.37±4.10 vs. 13.72±6.28, p &amp;lt; 0.005). At ≥ 13hpi, tPNf-tPNa was shorter in blastocysts than in cases of cleavage arrest (10.68±5.74 vs. 15.28±5.97, p &amp;lt; 0.05). It was particularly observed that tPNf-tPNa of GQ-BL in ≥ 13hpi was 7.90hpi, and t2 was not significantly different from blastocysts and GQ-BL in &amp;lt; 13hpi (25.06 vs. 25.66±3.18 vs. 25.19±2.97). Limitations, reasons for caution More data are needed for conclusive pregnancy results due to the small number of samples. Additionally, the maturation of oocytes had yet to be considered; further detailed studies related to oocyte maturity are needed. Wider implications of the findings Embryos with a tPNa of &amp;lt; 13hpi develop into GQ-BL. Embryos with delayed tPNa tend to overcome this by shortening the tPNf-tPNa, resulting in develop into blastocysts. Predicting blastocyst development can be achieved by considering factors such as tPNa∼t2. This can aid in improving the selection of embryos for cleavage-stage ET. Trial registration number N/A

O-261 Frozen gametes differ in post-insemination dynamics to their fresh sibling counterparts in cycles using testicular sperm

Abstract Study question Are there differences in post-insemination events between fresh and frozen-derived gametes when testicular sperm extraction (TESE) is used? Summary answer Frozen-TESE leads to decreased fertilization, but not blastocyst euploidy rates; 2PN-lag time is significantly longer when frozen gametes are utilized compared to their fresh counterparts What is known already Studies indicate that the origin of testicular sperm influences the development potential of cleavage-stage embryos into blastocysts. Recent morphokinetic analysis suggest that embryos from frozen-TESE exhibit prolonged pronuclear stages and more frequent uneven cleavage compared to those from ejaculated sperm. Notably, no comparison with fresh-TESE was performed. To date, no studies have explored analysis in embryo morphokinetic parameters comparing fresh and frozen-TESE, especially when derived from the same patient.Such comparisons could provide a more accurate understanding of potential differences. Given that TESE occasionally involves the use of vitrified oocytes, it is essential to thoroughly examine outcomes across all combinations. Study design, size, duration Retrospective cohort study in a tertiary referral IVF centre including 72 cycles of 30 couples between January 2017 to September 2023. Following insemination by TESE-sperm, embryos were cultured in time-lapse (TL) incubator. Blastocysts ≥BL3CC (Gardner’s) underwent trophectoderm (TE) biopsy for Preimplantation Genetic Testing for aneuploidy (PGT-A) by next generation sequency (NGS). Participants/materials, setting, methods Gamete sources were fresh (fresh-TESE) or frozen-thawed-TESE (frozen-TESE) from the same patients with non-obstructive azoospermia (NOA), and fresh or vitrified (frozen) oocytes from the same female partner. Fertilization, usable blastocysts, and euploidy rates were recorded. Morphokinetics were annotated for time (t) of PB2 (tPB2), tPNa (PN-appearance), tPNf (PN-fading), t2-t9, CP (compaction), tM (Morula), tSB (start blastulation), tB (blastocyst) and timings between each developmental event. Main results and the role of chance Among included 980 oocyte/TESE dyads, 253 (25.8%) used fresh oocyte/sperm, 141 (14.4%) frozen oocyte/fresh sperm, 558 (56.9%) fresh oocyte/frozen sperm and 28 (2.9%) frozen oocyte/sperm. In oocyte/sperm dyads using fresh-TESE, normal fertilization (2PN) (52.0% vs. 44.5%) were higher compared to dyads using frozen-TESE (P &amp;lt; 0.01). Mixed-effects logistic-regression accounting for dependency structure between oocytes/sperm dyads of the same couple, showed an association of frozen-TESE with lower fertilization (OR: 0.64, 95%CI: 0.45-0.89, P = 0.008) and lower blastulation (³BL3CC)/MII (OR: 0.53, 95%CI: 0.37-0.77, P = 0.001), but similar blastulation (³BL3CC)/2PN (OR: 0.69, 95%CI: 0.45-1.07, P = 0.09). After accounting for female age, euploidy per tested blastocyst was similar whether fresh or frozen sperm was used (OR: 0.91, 95% CI: 0.46-1.79, P = 0.788). Frozen oocyte use on the other hand was not significantly associated with these outcomes (P &amp;gt; 0.05 for all). Analysis of morphokinetic parameters revealed that use of frozen dyads was associated with longer duration of 2PN after tPNa to tPNf: in frozen-TESE, 2.10h longer (95%CI: 0.55-3.64, P = 0.008); and in frozen oocytes, 2.34h longer (95%CI: 0.31-4.36, P = 0.024). Among embryos reaching blastulation, frozen-TESE embryos reached 6-cell stage earlier (tPNf to t6, 2.4h earlier, P = 0.009) but they were somewhat delayed in reaching expansion from there on (t6 to tEB, 4.4 hours later, P = 0.059). Limitations, reasons for caution In certain oocytes, insemination was conducted using either fresh or frozen immotile sperm that showed unresponsiveness to activation methods. However, the samples size hinders a comprehensive comparison. Wider implications of the findings Use of frozen-thawed TESE from NOA-patients may decrease fertilization and consequently lower blastocyst yield, but not euploidy rates. Synchronizing fresh-TESE with oocyte retrieval appears safer to maximize fertilization. Regardless of the gamete type, frozen-sourced sperm or oocyte impact 2PN-lag time, possibly indicating nuclear repair mechanisms post-freezing, requiring further investigation. Trial registration number 2310-ABU-020-VF

P-204 Automated embryo assessment: a simple web application correlates scores with embryo morphology, developmental velocity, ploidy, and clinical outcomes

Abstract Study question How does the novel automated scoring system in our laboratory correlate with treatment success in different subpopulations? Summary answer The web app automates embryo assessment, linking scores to morphology, development speed, ploidy, and clinical outcomes in treatments with both patient and donor oocytes. What is known already In the realm of assisted reproductive technologies, numerous platforms employing artificial intelligence (AI) have emerged for embryo assessment. Substantial research has been dedicated to evaluating the efficacy of AI in predicting pregnancy outcomes, particularly following blastocyst transfer. However, not all available options have undergone external validation, a crucial step deemed necessary before investing in such tools. This project endeavors to validate the utility of EMBRYOAID v1.0, a straightforward web application capable of assessing embryos through photos and/or videos. Study design, size, duration This is a retrospective cohort study including 485 patients who underwent IVF treatments with embryos (n = 5,710) cultured in EmbryoScope® time-lapse systems. Senior embryologists routinely assessed blastocysts using ASEBIR morphological criteria (A-D). Then, time-lapse videos were scored by the EMBRYOAID algorithm (0-10). We investigated the relationship between automated scoring and: (I) conventional morphology, (II) morphokinetic parameters, (III) euploidy rate, and (IV) implantation in cycles with own oocytes, egg donation, and preimplantation genetic testing for aneuploidies (PGT-A). Participants/materials, setting, methods The association with morphology and morphokinetics was investigated in 5,710 embryos. Ploidy correlation involved 258 blastocysts undergoing trophectoderm biopsy and next-generation sequencing analysis. Implantation association was studied in 381 single blastocyst transfers. Finally, a multivariable analysis considered oocyte age, body mass index, type of transfer (fresh-frozen) and day of transfer (5-6) to quantify the relevance of automated scoring in different treatment subpopulations. Model performance was assessed through ROC curves and area under the curve calculations. Main results and the role of chance The results that involved grouping were quartiled based on sample size, calculated by the statistical software. This deep-learning score was related to conventional morphology (7.9±1 for A, n = 295; 6.9±1.3 for B n = 1168; 5.9±1.3 for C, n = 830; and 3.1±1.8 for D, n = 3,417)*. Embryos with faster development in early stages received higher scores, considering division times at 2, 3, 4, and 5 cells*. Euploidy rate increased as the embryo score was higher: 39.1% for ≤5.6 (n = 54), 45.5% for 5.6-6.2 (n = 61), 48.5% for 6.2-6.9 (n = 65) and 58.2% for &amp;gt;6.9 (n = 78). Similarly, the implantation rate was also higher in accordance with the score: 42.7% for ≤6.2 (n = 73), 50.6% for 6.2-7.4 (n = 87), 60.2% for 7.4-8.2 (n = 103) and 69.4% for &amp;gt;8.2 (n = 118). The stepwise multivariate analysis showed that embryo score was related to the odds of implantation in conventional treatments with patient oocytes (OR = 1.4; 95% CI [1.1-1.8]*) and in the oocyte donation program (OR = 1.2; 95% CI [1–1.4]*). The AUCs demonstrated the following performance: 0.643 (0.559-0.727) for patient oocytes and 0.677 (0,624-0,731) for the oocyte donation program. There was no significant association of embryo score with implantation in PGT-A treatments. Limitations, reasons for caution The retrospective nature of the study, despite serving as a thorough external validation, introduces inherent limitations. Notably, in PGT-A treatments, embryos underwent assisted hatching on day 3, potentially influencing the model’s performance. Wider implications of the findings The implications of our findings validate a user-friendly tool for automated embryo scoring in the laboratory. The achieved performance is comparable or superior to other tested models, particularly in egg donation programs, potentially encouraging clinics to adopt this tool routinely. Trial registration number Not applicable

O-246 A cutting-edge artificial intelligence system tracking embryo development and pinpointing the optimal day for blastocyst utilization

Abstract Study question Can an AI model predict at which specific day (D) of development a blastocyst will reach expansion at a usable stage for biopsy/cryopreservation/transfer? Summary answer Our model achieved a high accuracy tracking embryos’ developmental times while forecasting the day when they would reach a usable blastocyst stage. What is known already Forecasting the optimal day for blastocyst utilization is crucial. It can facilitate informed decisions on transfer/biopsy/cryopreservation, while streamlining the laboratory workflow. Previous prediction models on this task relied solely on embryo morphological and morphokinetic data. However, emerging approaches, incorporating computer image analysis, offer an opportunity to enhance their accuracy and objectivity. To our knowledge, none has specifically predicted the precise day of embryo development when the blastocyst has expanded enough to become usable. This information is invaluable to plan workload distribution on D5/6/7 of development. Moreover, with the rise of PGT-A, optimizing the time allocated for biopsy becomes essential. Study design, size, duration Retrospective study conducted at a tertiary IVF centre, utilizing data from June 2015 to January 2023. A total of 14,704 time-lapse videos (TLV) from embryos of 1,805 patients, cultured across 2,475 cycles, were analyzed split into train/validation/test by patient (to avoid data leaking) with 60/20/20 proportion. Outcome parameters encompassed the times for all embryo developmental stages or morphokinetic parameters, blastocyst formation and utilization, as well as the specific day of blastocyst utilization. Participants/materials, setting, methods Embryos from patients with primary/secondary infertility were included. Laboratory culture conditions followed normal clinical routine. All embryos were monitored using embryoscope time-lapse systems and annotated by experienced embryologists. Blastocysts graded ≥BL3CC (Gardner) were biopsied/cryopreserved/transferred on D5/6/7. Only videos depicting embryos with normal fertilization were considered in the model. Predictions for the utilization day were made 48-hours and 24-hours ahead of the blastocyst usable stage. Main results and the role of chance We developed a customized variant of the Temporal Fusion Transformer model, modified for analyzing time-sequence-images from TLV. This model architecture allows the analysis of the sequence and clinical variables, identifying its most influential aspects from the model’s perspective. In the first two tasks of blastocyst prediction and forecasting blastocyst utilization day, data was split into training-set: 8,802 TLV (1,083 patients); validation-set: 2,925 TLV (361 patients); and test-set: 2,977 TLV (361 patients). Our trained model achieved a strong performance on ‘early warning’ 48-hour ahead forecasting of embryo reaching a usable blastocyst stage. Utilizing only unlabelled time-lapse image observations of the embryo, it reached 90 AUROC score on the test-set. For ‘late warning’, 24-hour ahead, we reached 97 AUROC score on the test-set, almost ideal prediction. Finally, 91 AUROC score was achieved on the binary blastocyst stage classification task, just considering the blastocyst formation at any expansion. In the embryo morphokinetic stage-labelling task, data was split into training-set: 8,420 TLV (1,078 patients); validation-set: 2,777 TLV (360 patients); and test-set: 2,780 TLV (360 patients). We achieved competitive results with 95.5 macro-averaged AUROC score for multi-class (17 classes) prediction for each time-lapse-image. Each individual parameter from tPB2 to tHB reached a AUROC higher than 90. Limitations, reasons for caution The model needs to undergo an extensive cross-validation, to evaluate its performance, followed by a prospective validation. Wider implications of the findings This model belongs to the realm of IVF lab automation, offering substantial assistance in daily operations. It aids embryologists not only in annotating developmental stages with high accuracy, but also predicts the blastocyst utilization day, thereby providing crucial support for efficient workload planning and streamlined operations. Trial registration number NA