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.

[Repeated in vitro fertilization failure: Abnormalities identified in the diagnostic assessment]. / Échecs répétés de fécondation in vitro : anomalies retrouvées sur le bilan diagnostique.

OBJECTIVES: Investigate the proportion of abnormalities identified on the diagnostic assessment performed after at least two previous failed IVF attempts. Discuss the real benefit of this evaluation. METHODS: Retrospective descriptive study. Between January 2008 and January 2012, 205 couples with at least two consecutive failed IVF attempts had a diagnosis evaluation which consisted in couple's karyotypes; autoimmune and haemostasis biological check-up, pelvic ultrasound-Doppler and hysteroscopy for women. RESULTS: The main biological anomalies were autoimmune for 23.9% of women: antinuclear antibodies (5.7%), antithyroid peroxidase (11.5%) and antithyroglobulin (8.3%); thrombotic with antiphospholipid antibodies for 8.2% of women (1.4% lupus anticoagulant and 6.8% anticardiolipin antibodies), and heterozygous prothrombin gene mutation for 9.5%. Karyotypes were abnormal for 2.1% of women and 0% of men. Ultrasound-Doppler appeared to be abnormal in 44.7% of cases (pulsatility index of uterine artery≥3 and/or protodiastolic notch), and diagnostic hysteroscopy was abnormal in 14.6% of cases. In order to target the real implantation failure, we compared the groups "<8 embryos transferred" versus "≥8 embryos transferred" and "pregnancy after the third or fourth IVF cycle" versus "no pregnancy", but no statistically significant difference was found. CONCLUSION: The diagnostic assessment carried out for recurrent IVF failure can detect biological, karyotypic and morphological abnormalities, in the same proportion that in previous studies. Further studies will have to be conducted to evaluate the real impact of these abnormalities in the recurrent implantation failure and the effectiveness of therapeutic care.

[Thromboembolism and ovarian hyperstimulation]. / Accidents thrombo-emboliques et hyperstimulation ovarienne.

Three new cases of thromboembolic accidents involving the superior vena cave were associated with ovarian hyperstimulation and multiple pregnancy. On the basis of a review of the literature, three syndromes were individualized. First embolic arteries are infrequent and serious with an important morbidity; low doses of gonadotrophines is sufficient for causing this pathology. Second, thrombosis of the superior vena cava is late, during in vitro fertilization regimens with GnRH analogs complicated with serious hyperstimulation (degree III). Third, pregnancy is always present; pulmonary complications are very uncommon. Thrombosis of inferior vena cava is late, pregnancy is often present but women are frequently at risk of thromboembolic accident. Prevention of hyperstimulation is important. When present, HBPM is recommended for prevention of thromboembolism. Diagnostic of thrombosis is made with clinical examination and morphologic complementary exams such colour-vascular Doppler and magnetic resonance imaging. Curative treatment requires heparinization.

[The value of hysteroscopy after repeated implantation failures with in vitro fertilization]. / Intérêt de l'hystéroscopie après échecs répétés d'implantation en fécondation in vitro.

OBJECT: we attempt to determine the usefulness of uterine re-assessment by hysteroscopy in women with two unsuccessful IVF-ET attempts. MATERIAL AND METHOD: This retrospective study concerns seventy-three infertile women who have had a repeat uterine hysteroscopy after two implantation failures in IVF. RESULTS: In half the cases, an abnormality was diagnosed. Cervical abnormalities (synechia, polyp, false passage) and hormono-dependent abnormalities (polyp, hyperplasia, submucous myoma) were the most frequent findings. The rate of pregnant women raises to twenty-two percent after treatment of these abnormalities. CONCLUSION: Regarding our experience, it seems legitimate to perform hysteroscopy in women who have had two IVF-ET failures before attempting a third procedure.

[Can the type of GnRH used during long in vitro fertilization protocols influence the pregnancy rate?]. / Le type d'analogue de la GnRH utilisé au cours des protocoles longs en FIV peut-il influencer les taux de grossesse?

Luteal defect is common during IVF cycles using GnRH agonist. It could be interesting to make up for this problem by using short acting GnRH agonist (effective for 24 hours), more handy at the end of the stimulation. 160 patients included in a long IVF protocol at the Bordeaux CHRU FIV center have been randomized, from September 1993 to August 1994, for the analog's choice (long or short) at the beginning of the stimulation. The stimulation's parameters, the hormonal dosages and the pregnancy rate are independent of the galenic form used. In close, the use of long-acting GnRH analog (sympler to use) sems preferable.