A hybrid artificial intelligence model leverages multi-centric clinical data to improve fetal heart rate pregnancy prediction across time-lapse systems.

STUDY QUESTION: Can artificial intelligence (AI) algorithms developed to assist embryologists in evaluating embryo morphokinetics be enriched with multi-centric clinical data to better predict clinical pregnancy outcome? SUMMARY ANSWER: Training algorithms on multi-centric clinical data significantly increased AUC compared to algorithms that only analyzed the time-lapse system (TLS) videos. WHAT IS KNOWN ALREADY: Several AI-based algorithms have been developed to predict pregnancy, most of them based only on analysis of the time-lapse recording of embryo development. It remains unclear, however, whether considering numerous clinical features can improve the predictive performances of time-lapse based embryo evaluation. STUDY DESIGN, SIZE, DURATION: A dataset of 9986 embryos (95.60% known clinical pregnancy outcome, 32.47% frozen transfers) from 5226 patients from 14 European fertility centers (in two countries) recorded with three different TLS was used to train and validate the algorithms. A total of 31 clinical factors were collected. A separate test set (447 videos) was used to compare performances between embryologists and the algorithm. PARTICIPANTS/MATERIALS, SETTING, METHODS: Clinical pregnancy (defined as a pregnancy leading to a fetal heartbeat) outcome was first predicted using a 3D convolutional neural network that analyzed videos of the embryonic development up to 2 or 3 days of development (33% of the database) or up to 5 or 6 days of development (67% of the database). The output video score was then fed as input alongside clinical features to a gradient boosting algorithm that generated a second score corresponding to the hybrid model. AUC was computed across 7-fold of the validation dataset for both models. These predictions were compared to those of 13 senior embryologists made on the test dataset. MAIN RESULTS AND THE ROLE OF CHANCE: The average AUC of the hybrid model across all 7-fold was significantly higher than that of the video model (0.727 versus 0.684, respectively, P = 0.015; Wilcoxon test). A SHapley Additive exPlanations (SHAP) analysis of the hybrid model showed that the six first most important features to predict pregnancy were morphokinetics of the embryo (video score), oocyte age, total gonadotrophin dose intake, number of embryos generated, number of oocytes retrieved, and endometrium thickness. The hybrid model was shown to be superior to embryologists with respect to different metrics, including the balanced accuracy (P ≤ 0.003; Wilcoxon test). The likelihood of pregnancy was linearly linked to the hybrid score, with increasing odds ratio (maximum P-value = 0.001), demonstrating the ranking capacity of the model. Training individual hybrid models did not improve predictive performance. A clinic hold-out experiment was conducted and resulted in AUCs ranging between 0.63 and 0.73. Performance of the hybrid model did not vary between TLS or between subgroups of embryos transferred at different days of embryonic development. The hybrid model did fare better for patients older than 35 years (P < 0.001; Mann-Whitney test), and for fresh transfers (P < 0.001; Mann-Whitney test). LIMITATIONS, REASONS FOR CAUTION: Participant centers were located in two countries, thus limiting the generalization of our conclusion to wider subpopulations of patients. Not all clinical features were available for all embryos, thus limiting the performances of the hybrid model in some instances. WIDER IMPLICATIONS OF THE FINDINGS: Our study suggests that considering clinical data improves pregnancy predictive performances and that there is no need to retrain algorithms at the clinic level unless they follow strikingly different practices. This study characterizes a versatile AI algorithm with similar performance on different time-lapse microscopes and on embryos transferred at different development stages. It can also help with patients of different ages and protocols used but with varying performances, presumably because the task of predicting fetal heartbeat becomes more or less hard depending on the clinical context. This AI model can be made widely available and can help embryologists in a wide range of clinical scenarios to standardize their practices. STUDY FUNDING/COMPETING INTEREST(S): Funding for the study was provided by ImVitro with grant funding received in part from BPIFrance (Bourse French Tech Emergence (DOS0106572/00), Paris Innovation Amorçage (DOS0132841/00), and Aide au Développement DeepTech (DOS0152872/00)). A.B.-C. is a co-owner of, and holds stocks in, ImVitro SAS. A.B.-C. and F.D.M. hold a patent for 'Devices and processes for machine learning prediction of in vitro fertilization' (EP20305914.2). A.D., N.D., M.M.F., and F.D.M. are or have been employees of ImVitro and have been granted stock options. X.P.-V. has been paid as a consultant to ImVitro and has been granted stocks options of ImVitro. L.C.-D. and C.G.-S. have undertaken paid consultancy for ImVitro SAS. The remaining authors have no conflicts to declare. TRIAL REGISTRATION NUMBER: N/A.

Is it worth it to cryopreserve embryos with blastulation delay at day 5?

This is a retrospective study over a 5-year period. In total, 3139 embryos were individually cryopreserved (Cryotop®) and warmed using the Kitazato vitrification/warming kit. They were classified into three categories based on their expansion degree. Transfer, implantation and pregnancy rates were assessed for each embryo category and compared using SPSS (Statistical Package for the Social Sciences) software. In total, 1139 couples enrolled in infertility treatment programme benefitted from embryo vitrification at day 5. After warming, embryos belonging to the three categories showed similar success rates. Although there was a trend towards better outcomes when grade 3 embryos were transferred, the differences did not reach statistical significance: implantation rates (n fetal sac/n embryo transferred) grade 1: 21.9%, grade 2: 22.7% and grade 3: 30.3% (=0.19). Pregnancy rate (n clinical pregnancy/n transfer) (21.9%, 22.7%, 30.3%, respectively; P=0.11). Miscarriage rate was not statistically different in the three categories (14.5%, 20.4%, 20%, respectively, P=0.51). Our overall results show that it is worth vitrifying slow kinetics embryos as they provide a non-negligible chance to give rise to a pregnancy.

Morphokinetics analysis of embryos derived from vitrified/warmed oocytes.

PURPOSE: Oocyte vitrification is a worldwide used technique that has proved its worth. Although it was shown not to alter oocyte integrity, a recent study concluded that it may affect oocyte embryo development. As the morphology and kinetics of embryos derived from sibling fresh and vitrified oocytes have not been described previously, the present study evaluates cleavage rate, blastomeres size, fragmentation rate, and blastocyst formation in vitrified/warmed oocyte derived embryos (VODE) as compared with sibling fresh oocytes derived embryos (FODE). METHODS: This investigation included 90 infertility cases displaying large cohort of mature oocytes at pick up, divided into 2 groups after denudation. A part of oocytes underwent ICSI while others were vitrified. Oocyte warming cycles were performed when no pregnancy was achieved using fresh eggs. Zygote to blastocyst development was recorded prospectively in an image database up to day 5. RESULTS: VODE did not show major difference as compared with FODE in terms of cleavage rate, number of blastomeres, fragmentation rate, and blastomeres size. Furthermore, percentage of morulae at day 4 and blastocysts at day 5 are not affected by oocyte vitrification. CONCLUSION: Although our results show that embryo development is not altered by oocyte vitrification, offspring follow-up is essential to exclude any adverse developmental effect of the technique.

Normal live birth after vitrified/warmed oocytes intracytoplasmic sperm injection with immotile spermatozoa in a patient with Kartagener's syndrome.

The present article is a report on two cases of male Kartagener's syndrome enrolled in intraconjugal IVF programme due to akinetospermia. Viable spermatozoa were selected using a hypo-osmotic swelling test (HOST) and pentoxifylline activation and subsequently microinjected into vitrified/warmed oocytes. The treatment enabled one of these two couples to achieve a pregnancy and to give birth to a healthy baby girl.

[Sex chromosomes and meiosis]. / Chromosomes sexuels et méiose.

Sex chromosome behaviour fundamentally differs between male and female meiosis. In oocyte, X chromosomes synapse giving a XX bivalent which is not recognizable in their morphology and behaviour from autosomal bivalents. In human male, X and Y chromosomes differ from one another in their morphology and their genetic content, leading to a limited pairing and preventing genetic recombination, excepted in homologous region PAR1. During pachytene stage of the first meiotic prophase, X and Y chromosomes undergo a progressive condensation and form a transcriptionally silenced peripheral XY body. The condensation of the XY bivalent during pachytene stage led us to describe four pachytene substages and to localize the pachytene checkpoint between substages 2 and 3. We also defined the pachytene index (PI=P1+P2/P1+P2+P3+P4) which is always less than 0.50 in normal meiosis. XY body undergoes decondensation at diplotene stage, but transcriptional inactivation of the two sex chromosomes or Meiotic Sex Chromosome Inactivation (MSCI) persists through to the end of spermatogenesis. Sex chromosome inactivation involves several proteins, some of them were now identified. Two isoforms of the HP1 protein, HP1beta and HP1gamma, are involved in the facultative heterochromatinization of the XY body, but the initiation of this process involves the phosphorylation of the protein H2AX by the kinase ATR whose recruitment depends on BRCA1. Extensive researches on the inactivation of the sex chromosomes during male meiosis will allow to a better understanding of some male infertilities.

[Macrocephalic spermatozoa. What would be the impact on reproduction?]. / Les spermatozoïdes macrocéphales. Quels risques pour la fonction de reproduction ?

OBJECTIVE: We want to highlight the risk of infertility and failure of Assisted Reproductive Technologies due to the presence of macrocephalic spermatozoa (MS) in the sperm at rate equalling or superior to 20% in at least one semen analysis. PATIENTS AND METHODS: We did a retrospective analysis of 19 infertile patients presenting MS at average rate between 14.3 and 49.7%. For each patient, at least one semen analysis showed a MS rate equal or superior to 20%. We did an automated analysis of the spermatozoa surface for 13 patients and a detailed analysis of the MS morphology in 18 patients. Thirteen couples benefited of one or more IVF with or without ICSI. RESULTS: The semen analysis shows an impairment of one or more parameter of the sperm in all patients. Three morphological aspects for MS were highlighted: MS with irregular head, MS with regular head, and MS with multiple heads, with a dominance of irregular heads. The spermatozoa surface analysis shows a significant increase of the average surface and of the standard deviation (p<0.0001). The average rate of pregnancies by transfer is decreased compared to usual rates in our laboratories (13% versus 28%). DISCUSSION AND CONCLUSION: We want to sensitize biologist and clinical doctors to the existence of partial forms of this syndrome, which could be related to infertility with impaired sperm parameters and low pregnancy rates after FIV or ICSI.

[Genetic aspects of the teratozoospermia]. / Aspects génétiques de la tératozoospermie.

Recent mutation identification in well-known sperm defects gives proof that there are genetic causes of infertility. Familial forms and some features of the spermograms lead toward the genetic origin of these syndromes. For each syndrome, several clinical aspects and partial forms were described. In these latter, apparently normal spermatozoa coexist with those showing the phenotype of interest. Transmission electron microscopy is the better tool to characterize the specific details of each syndrome. The frequency of genetic teratozoospermia is weak, the most studied syndromes are the globozoospermia, the macrocephaly, the syndrome of decapitated spermatozoa and the dyplasia of the fibrous sheat. A mutation was identified for two from these syndromes, but the two mutations does not account for all the cases from each syndrome. The various clinical aspects observed for each syndrome suggest that either other mutations or other genes are probably involved in these spermatogenic failures. The use of spermatozoa from patients for intra cytoplasmic sperm injection (ICSI) may pose two problems: fertilization problems and genetic risk for the progeny, including chromosomic and genic risk. Except for total macrocephaly which is excluded from ICSI because of sperm chromosomal abnormalities, these syndromes are consistent with assisted fertilization, but with uncertain rates of fertilization and pregnancy.

[The hand: embryology and main malformative mechanisms]. / La main: embryologie et principaux mécanismes malformatifs.

Upper limb bud appears in the cervical region of the embryo during the fifth week of development. It is made of epithelia and underlying mesenchyme. Diffusible growth factors, expressed by the apical ectodermal ridge, direct the proximal-distal growth. Other factors are expressed by zone of polarizing activity and ectoderm. They induce together anterior-posterior growth and dorsal-ventral polarity of the limb bud. The development of axial skeleton pattern is controlled by transcription factors from the HOX family, which are expressed in a stripe along the proximal and distal edges of the limb bud. Embryologic mechanisms of the main hand malformations are described, as well as their known genetic or mechanical aetiologies.

Meiotic abnormalities in patients bearing complete AZFc deletion of Y chromosome.

BACKGROUND: We studied meiosis in three infertile patients presenting complete AZFc microdeletion and three controls. METHODS: Primary spermatocytes were immunolabeled with SCP3, BRCA1 and gammaH2AX. We quantified the leptotene, zygotene and pachytene stages, and pachytene abnormalities: asynapsis and fragmented and dotted synaptonemal complexes (SCs). RESULTS: SCP3 level was significantly higher in leptotene and zygotene (bouquet) stages in patients, suggesting AZFc may have a direct effect on early prophase. SCs were abnormal in 77.3% of pachytene nuclei of patients versus 30.8% of controls. The two groups differed significantly (P < 0.001) in asynapsed nuclei, fragmented SC and dotted SCs. In patients, asynapsis were short and limited to a few bivalents. Staging of pachytene nuclei based on the morphology of the XY pair with BRCA1 revealed a prevalence of early pachytene substages (70.7%) in patients. H2AX was normally phosphorylated. CONCLUSIONS: In the absence of the AZFc region, the transient zygotene stage is extended, and chromosome condensation is reduced. The low level of limited asynapsis, the normal H2AX staining and the incomplete loss of germ cells at the pachytene checkpoint indicate that the AZFc region is not critical for meiotic recombination. We suggest that the pachytene phenotype develops secondarily to a primary defect that influences meiosis.