BlastAssist: a deep learning pipeline to measure interpretable features of human embryos.

STUDY QUESTION: Can the BlastAssist deep learning pipeline perform comparably to or outperform human experts and embryologists at measuring interpretable, clinically relevant features of human embryos in IVF? SUMMARY ANSWER: The BlastAssist pipeline can measure a comprehensive set of interpretable features of human embryos and either outperform or perform comparably to embryologists and human experts in measuring these features. WHAT IS KNOWN ALREADY: Some studies have applied deep learning and developed 'black-box' algorithms to predict embryo viability directly from microscope images and videos but these lack interpretability and generalizability. Other studies have developed deep learning networks to measure individual features of embryos but fail to conduct careful comparisons to embryologists' performance, which are fundamental to demonstrate the network's effectiveness. STUDY DESIGN, SIZE, DURATION: We applied the BlastAssist pipeline to 67 043 973 images (32 939 embryos) recorded in the IVF lab from 2012 to 2017 in Tel Aviv Sourasky Medical Center. We first compared the pipeline measurements of individual images/embryos to manual measurements by human experts for sets of features, including: (i) fertilization status (n = 207 embryos), (ii) cell symmetry (n = 109 embryos), (iii) degree of fragmentation (n = 6664 images), and (iv) developmental timing (n = 21 036 images). We then conducted detailed comparisons between pipeline outputs and annotations made by embryologists during routine treatments for features, including: (i) fertilization status (n = 18 922 embryos), (ii) pronuclei (PN) fade time (n = 13 781 embryos), (iii) degree of fragmentation on Day 2 (n = 11 582 embryos), and (iv) time of blastulation (n = 3266 embryos). In addition, we compared the pipeline outputs to the implantation results of 723 single embryo transfer (SET) cycles, and to the live birth results of 3421 embryos transferred in 1801 cycles. PARTICIPANTS/MATERIALS, SETTING, METHODS: In addition to EmbryoScope™ image data, manual embryo grading and annotations, and electronic health record (EHR) data on treatment outcomes were also included. We integrated the deep learning networks we developed for individual features to construct the BlastAssist pipeline. Pearson's χ2 test was used to evaluate the statistical independence of individual features and implantation success. Bayesian statistics was used to evaluate the association of the probability of an embryo resulting in live birth to BlastAssist inputs. MAIN RESULTS AND THE ROLE OF CHANCE: The BlastAssist pipeline integrates five deep learning networks and measures comprehensive, interpretable, and quantitative features in clinical IVF. The pipeline performs similarly or better than manual measurements. For fertilization status, the network performs with very good parameters of specificity and sensitivity (area under the receiver operating characteristics (AUROC) 0.84-0.94). For symmetry score, the pipeline performs comparably to the human expert at both 2-cell (r = 0.71 ± 0.06) and 4-cell stages (r = 0.77 ± 0.07). For degree of fragmentation, the pipeline (acc = 69.4%) slightly under-performs compared to human experts (acc = 73.8%). For developmental timing, the pipeline (acc = 90.0%) performs similarly to human experts (acc = 91.4%). There is also strong agreement between pipeline outputs and annotations made by embryologists during routine treatments. For fertilization status, the pipeline and embryologists strongly agree (acc = 79.6%), and there is strong correlation between the two measurements (r = 0.683). For degree of fragmentation, the pipeline and embryologists mostly agree (acc = 55.4%), and there is also strong correlation between the two measurements (r = 0.648). For both PN fade time (r = 0.787) and time of blastulation (r = 0.887), there's strong correlation between the pipeline and embryologists. For SET cycles, 2-cell time (P < 0.01) and 2-cell symmetry (P < 0.03) are significantly correlated with implantation success rate, while other features showed correlations with implantation success without statistical significance. In addition, 2-cell time (P < 5 × 10-11), PN fade time (P < 5 × 10-10), degree of fragmentation on Day 3 (P < 5 × 10-4), and 2-cell symmetry (P < 5 × 10-3) showed statistically significant correlation with the probability of the transferred embryo resulting in live birth. LIMITATIONS, REASONS FOR CAUTION: We have not tested the BlastAssist pipeline on data from other clinics or other time-lapse microscopy (TLM) systems. The association study we conducted with live birth results do not take into account confounding variables, which will be necessary to construct an embryo selection algorithm. Randomized controlled trials (RCT) will be necessary to determine whether the pipeline can improve success rates in clinical IVF. WIDER IMPLICATIONS OF THE FINDINGS: BlastAssist provides a comprehensive and holistic means of evaluating human embryos. Instead of using a black-box algorithm, BlastAssist outputs meaningful measurements of embryos that can be interpreted and corroborated by embryologists, which is crucial in clinical decision making. Furthermore, the unprecedentedly large dataset generated by BlastAssist measurements can be used as a powerful resource for further research in human embryology and IVF. STUDY FUNDING/COMPETING INTEREST(S): This work was supported by Harvard Quantitative Biology Initiative, the NSF-Simons Center for Mathematical and Statistical Analysis of Biology at Harvard (award number 1764269), the National Institute of Heath (award number R01HD104969), the Perelson Fund, and the Sagol fund for embryos and stem cells as part of the Sagol Network. The authors declare no competing interests. TRIAL REGISTRATION NUMBER: Not applicable.

Oocyte retrieval in social fertility preservation: challenging traditional beliefs on gonadotropin dosing and follicular size.

OBJECTIVES: To investigate treatment approaches for fertility preservation patients, with a focus on timing of oocyte retrieval, and to determine whether their characteristics differ from those of other IVF patients. Additionally, to evaluate the significance of follicle size on triggering day in the context of fertility preservation. METHODS: This retrospective cohort study was conducted in a tertiary, university-affiliated medical center. It compared 140 matched patients undergoing social fertility preservation to 140 patients undergoing IVF treatment due to male factor infertility. RESULTS: Patients undergoing fertility preservation received a higher initial gonadotropin dose and had more oocytes retrieved than the control group. Within the fertility preservation cohort, a negative correlation was observed between the rate of large follicles and the number of retrieved oocytes. While there was no significant association between rate of large follicles and oocyte maturation rate in the entire group, age-stratified analysis revealed a negative relationship. Analysis revealed that although traditional treatment determinants such as follicular size and gonadotropin dosing were considered, peak estradiol levels were consistently identified as significant predictors of treatment outcomes. CONCLUSIONS: Physicians may modify treatments for fertility preservation, emphasizing a higher gonadotropin dosage to maximize oocyte retrieval. Elevated estradiol levels can serve as a real-time predictive marker for the number of mature oocytes. While treatment strategies can influence outcomes, intrinsic patient factors, particularly baseline ovarian function, remain crucial. These results challenge beliefs regarding the importance of larger follicles and suggest the need for a tailored approach, considering patient age and specific fertility preservation objectives.

A clinical predictive model for live birth in women of advanced age undergoing PGT cycles.

PURPOSE: The trend of delaying childbirth has resulted in a growing number of advanced-aged women who are opting for preimplantation genetic testing (PGT) to screen for monogenic diseases or structural chromosomal rearrangements (PGT-M and PGT-SR). This increase in demand necessitates the development of a clinical predictive model for live birth outcomes in these women. Therefore, the objective of this study is to construct a comprehensive predictive model that assesses the likelihood of achieving a successful live birth in advanced-aged women undergoing PGT-M and PGT-SR treatments. METHODS: A retrospective cohort study of 37-45-year-old women undergoing preimplantation genetic testing for monogenic disease or structural chromosomal rearrangement cycles from 2010 to 2021 was conducted at a university hospital reproductive centre. The purpose was to develop a clinical predictive model for live birth in these women. The main outcome studied was the cumulative live birth rate in the first or subsequent cycles. Developing a decision tree enabled a comprehensive study of clinical parameters and expected outcomes. RESULTS: The analysis included 158 women undergoing 753 preimplantation genetic testing cycles. The cumulative live birth rate was 37.342% (59/158). Decision tree analysis revealed that women aged ≤ 40.1 or women > 40.1 with one or more top-quality transferable embryos in their first cycle had the best chance for a live baby (56% and 41%, respectively). Those older than 40.1 without top-quality embryos and seven or fewer dominant follicles had no live births. A Kaplan-Meier curve showed that for autosomal dominant diseases, there was a negligible increase in live birth rate after three cycles, compared to six cycles in autosomal recessive inheritance. CONCLUSION: In older women, the chance of delivering after repeated cycles is higher in those with at least one top-quality unaffected embryo in their first preimplantation genetic testing cycle. Additional preimplantation genetic testing cycles after three in carriers of an autosomal dominant disorder and six in those with an autosomal recessive disorder should be considered prudently.

Does the SARS-CoV-2 mRNA vaccine and its serum IgG levels affect fertility treatments and obstetric outcomes? An observational cohort study.

BACKGROUND: Although there are some data regarding the COVID-19 vaccine and in in vitro fertilization (IVF) treatments, its potential impact in terms of serum immunoglobulin G (IgG) levels has not been evaluated prospectively. This study aimed to assess the effect of COVID-19 vaccine and IgG levels on IVF outcomes. METHODS: This observational, cohort study was conducted at a referral IVF unit. Couples undergoing IVF treatment during the COVID-19 vaccination period were recruited from March-April 2021. The study compared 38 women who had received the Pfizer mRNA COVID-19 vaccination to 10 women who had not and were not infected by the virus. We also compared pre- and post-vaccination IVF treatments for 24 women. The relation between serologic titers and IVF treatment outcomes was also assessed. RESULTS: No significant difference was found between the vaccinated and unvaccinated/uninfected groups regarding the main outcome measures. However, there was a trend toward a higher pregnancy rate for the unvaccinated group (57% vs. 23%, p = 0.078) but no difference in delivery rate (p = 0.236), gestational week (p = 0.537) or birth rate (p = 0.671). CONCLUSION: We cautiously state that the COVID-19 mRNA vaccine does not affect fertility outcomes, including fertilization, pregnancy and delivery rates, obstetric outcomes, and semen parameters, regardless of measured IgG levels.

Cleavage stage at compaction-a good predictor for IVF outcome.

OBJECTIVE: To analyze whether cleavage stage at compaction, and not only kinetics, can serve as a reliable predictor for clinical outcome. METHODS: A retrospective cohort study including 1194 embryos, classified by compaction initiation stage (Group 1: compaction at fewer than eight cells, Group 2: compaction at eight cells, Group 3: compaction at more than eight cells). Of these, 815 embryos were evaluated for morphokinetic preimplantation parameters, and 379 embryos were analyzed for clinical implantation following thawing and transfer of single blastocysts during the same period. RESULTS: In total, 1194 embryos were analyzed. Embryos that underwent compaction from more than eight cells (Group 3) exhibited more synchronous cleavage compared with Groups 1 and 2 (at both S2 and S3; P < 0.001), and displayed a significantly lower fragmentation rate. The likelihood of obtaining top-quality blastocysts decreased by 73% and 44% when comparing Group 3 embryos with those of Groups 1 and 2, respectively, (P < 0.03). Clinical validation of the results shows that while compaction from fewer than eight cells barely produced blastocysts for transfer, compaction at eight or more cells is crucial for implantation and birth (birth rates 11.1% and 18.5% for Groups 2 and 3, respectively). CONCLUSION: Cleavage stage at compaction has a direct effect on blastocyst quality and subsequent pregnancy, so can be included in newly developed deep learning models for embryo selection.

General anesthesia with propofol during oocyte retrieval and in vitro fertilization outcomes: retrospective cohort study.

General anesthesia is frequently administered during oocyte retrieval. Its effects on the outcomes of IVF cycles are uncertain. This study investigated whether administration of general anesthesia (specifically propofol) during oocyte retrieval affects IVF outcomes. A total of 245 women undergoing IVF cycles were included in this retrospective cohort study. IVF outcomes of 129 women who underwent oocyte retrieval under propofol anesthesia and 116 without anesthesia were compared. Data were adjusted for age, BMI, estradiol on triggering day and total gonadotropin dose. The primary outcomes were fertilization, pregnancy and live birth rates. A secondary outcome was the efficiency of follicle retrieval associated with the use of anesthesia. Fertilization rate was lower in retrievals under anesthesia compared to without (53.4% ± 34.8 vs. 63.7% ± 33.6, respectively; p = 0.02). There was no significant difference in the ratio of expected to retrieved oocytes between retrievals with and without anesthesia (0.8 ± 0.4 vs. 0.8 ± 0.8, respectively, p = 0.96). The differences in pregnancy and live birth rates between the groups were not statistically significant. General anesthesia administered during oocyte retrieval may have adverse effects on the fertilization potential of oocytes. This impact on the developmental potential of oocytes may lead to negative IVF outcomes and should be investigated further.

High physical activity and ovarian reserve: a prospective study of normo-ovulatory professional athletes.

BACKGROUND: This study investigated whether high physical activity affects ovarian reserve in normo-ovulatory, reproductive-age women. METHODS: This prospective, observational study compared 31 professional female athletes, with 31 women who did not engage in physical activity. It was conducted 2017-2020 in a tertiary medical center. Normo-ovulatory, professional athletes, ages 20-35 years were recruited from The Wingate Institute-the Israeli National Institute for Sport Excellence. They had high International Physical Activity Questionnaire (IPAQ) scores. Non-physically active women, matched by age and body mass index, were recruited from hospital staff. Women were evaluated for ovarian reserve markers on day 2-5 of the menstrual cycle, including follicular stimulating hormone, antral follicle count, anti-Mullerian hormone and Inhibin B. RESULTS: The average age of the high physical activity group was 29.9 ± 4.2 years and the nonactive group 31.6 ± 4.2 years (p = 0.062). Body mass index of both groups were similar (22.5 ± 5.0 vs. 21.4 ± 2.5, respectively; p = 0.1). No differences were observed with respect to follicle stimulating hormone (p = 0.12) and anti-Mullerian hormone (p = 0.16). A trend towards higher total antral follicle count in the high physical activity group vs. the non-active group (34.5 ± 12.9 vs. 28.1 ± 15.2, p = 0.08) and lower Inhibin B (68.1 ± 36.8 vs. 89.4 ± 46.1, p = 0.05). Menarche age correlated with anti-Mullerian hormone (r = 0.387, p = 0.003), as did total antral follicle count (r = 0.368, p = 0.004). IPAQ scores and basal follicle stimulating hormone levels were negatively correlated (r = - 0.292, p = 0.005). CONCLUSIONS: Athletic, normo-ovulatory women have ovarian reserves that are at least as good as those of the general population. As this is the first study examining this issue, it could cautiously reassure women engaged in high physical activity regarding ovarian reserve.

Blood fatty acid analysis reveals similar n-3 fatty acid composition in non-pregnant and pregnant women and their neonates in an Israeli pilot study.

Maternal docosahexaenoic acid (DHA) is required during pregnancy to supply for normal fetal growth and development. This pilot study aimed to assess the unknown fatty acid (FA) composition in a cohort of non-pregnant and pregnant Israeli women at term and their offspring on a normal diet without n-3 FA supplementation. The fatty acid profile, analyzed using gas chromatography, showed significantly higher plasma monounsaturated (MUFA) and lower n-6 FA percent distribution with similar n-3 index, in pregnant compared to non-pregnant women. RBC exhibited significantly higher MUFA with similar n-3 index, in pregnant compared to non-pregnant women. N-3 FA significantly correlated between neonates' plasma, with higher n-3 index, and pregnant women's DHA. Conclusion: DHA levels in non-pregnant and pregnant Israeli women at term were comparable and the DHA in pregnant women's plasma positively correlated with their neonate's level, suggesting an efficient mother-fetus FA transfer and/or fetal fatty acid metabolism to longer FA products.

Trial of labor after cesarean (TOLAC) in women with premature rupture of membranes.

Introduction: The aim of this study was to assess the success rate of a trial of labor after a previous cesarean section (TOLAC) in the settings of premature rupture of membranes (PROM) and to compare conservative management with spontaneous labor and induction of labor.Methods: This was a retrospective cohort study conducted in a single tertiary care center between January 2011 and March 2017. Women with singleton pregnancy and a previous cesarean section (CS) who presented with PROM and underwent TOLAC were included. Outcomes and rate of successful vaginal delivery after induction of labor were compared to conservative treatment and spontaneous labor.Results: Among 830 women who met the inclusion criteria, 723 (87.1%) had a spontaneous onset of labor following PROM and 107 (12.9%) had an induction of labor. The rate of successful TOLAC was similar between the groups (75.7 vs. 81.6%, respectively, p = .22). However, induction of labor was associated with an increased risk for uterine rupture (1.87 vs. 0.96%, p < .001), operative complications (6.7 vs. 2.3%, p < .001), and composite maternal postpartum complications (21.4 vs. 10.7%, respectively, p = .014) compared to conservative management with spontaneous initiation of labor. There was no difference in neonatal outcome between the groups.Conclusion: Induction of labor following PROM in women with a previous CS is associated with high successful vaginal delivery rate. However, the risk for uterine rupture and operative and maternal complications is significantly increased compared to spontaneous initiation of labor.

Morphokinetic characteristics of embryos originating from extremely small follicles: A prospective study.

OBJECTIVE: To investigate the developmental potential of oocytes and embryos derived from extremely small follicles (<10 mm) in comparison to those originated in larger follicles. STUDY DESIGN: A prospective study, undertaken in a university affiliated single center tertiary hospital. The study included 98 patients undergoing infertility treatments. On the day of ovum pickup (OPU) follicles were counted and measured. Aspiration of follicles larger and smaller than 10 mm was undertaken separately and the development of embryos originating from oocytes from these follicles was followed up using different wells for each embryo. There was no low limit of size for aspiration. Each oocyte retrieved was marked for its origin and numbered for further follow up. We recorded: Oocytes retrieved, maturation stage, fertilization rate, cleavage rate, morphokinetic parameters, embryo transfers, embryo freezing, oocyte freezing and biopsy rate for preimplantation genetic diagnosis (PGD). Quality was evaluated by the morphokinetic parameters of the embryos developed using time-lapse imaging technology. Day 3 KIDScore was calculated to all embryos. RESULTS: Small follicles compared to large follicles displayed lower recovery rate (45% vs. 74%, P < 0.0001), fewer matured oocytes (37.5% vs. 61.7%, P < 0.0001), higher rates of GV oocytes (20.7% vs., 3.7%, P < 0.0001), and lower fertilization rate (43.7% vs. 63.3%, P < 0.0001. However, morphokinetic variables were similar between embryos that originated from either small or large follicles. Median KIDscores were identical for embryos from small or large follicle origin. CONCLUSIONS: Embryos originated from small follicles were not different than embryos from larger follicles, as assessed by morphokinetic parameters in time lapse system. In view of our findings, physicians should bear in mind that small follicle aspiration might yield good quality embryos.

Characteristics and severity of preeclampsia in young and elderly gravidas with hypertensive disease.

INTRODUCTION: Advanced maternal age (AMA) is associated with increased risk for preeclampsia, however, a paucity of data exists regarding the characteristics of the disease in this age group. Our aim was to compare the characteristics and severity of preeclampsia in older and younger gravidas. MATERIALS & METHODS: A retrospective, small case control study of women diagnosed with preeclampsia in a single tertiary care center. Nulliparous women ≥40 years old with singleton pregnancies ≥ 24 0/7 weeks' gestation were matched (1:2 ratio) with young (20-34 years old) nulliparous women. RESULTS: The rate of severe preeclampsia (60.9 vs 69.6% respectively), HELLP, eclampsia or the need for magnesium treatment did not differ between the groups. However, the AMA group had an increased rate of postpartum presentation or exacerbation of preeclampsia compared to the control group (50.0 vs. 28.3% respectively, p = 0.01). In the AMA group, 93.5% of births were by cesarean section (CS) compared to 52.2% in the control group (p < 0.0001). There was no difference in birthweight, rate of small for gestational age or composite neonatal morbidity between the groups. CONCLUSIONS: Preeclampsia at an advanced maternal age carries a similar rate of severe preeclampsia and complications as in young women. However, women over 40 years old have an increased risk for presentation or exacerbation of preeclampsia in the postpartum period and an increased rate of CS compared to younger gravidas.