Search

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 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.

Study question Is the endogenous progesterone level (EPL) on the day of hCG administration associated with pregnancy outcomes in modified natural cycles before embryo transfer? Summary answer The present analysis suggested the negative effect of EPL on clinical pregnancy when progesterone was above 1,1 ng/ml on the day of HCG triggering. What is known already Endogenous progesterone levels (EPL) experience an abrupt increase following the ovulatory LH surge. It is described that this change in basal EPL contributes to modulating the endometrial receptivity in preparation for embryo implantation. Several authors reported a negative impact on the pregnancy of high serum EPL on the day of HCG. These studies revealed accelerated endometrial changes and a subsequent decoupling of the implantation window. These observations involved only patients who underwent ovarian stimulation before the embryo transfer. However, studies based on EPL in modified natural cycles before blastocyst transfer are very limited. Study design, size, duration The present non-interventional pilot cohort study aimed to determine the EPL in patients undergoing an endometrium preparation for frozen-thawed embryo transfer in non-stimulated cycles, and its association with clinical pregnancy rates. A total of 46 participants were included in this analysis. To reduce variability, the mean EPL was calculated excluding 2 outlier values in our dataset. The cut-off value of EPL was established using the Youden-index, and the T-student was used for further statistical comparisons. Participants/materials, setting, methods Patients following our IVF program, who underwent frozen-thawed embryo transfer during the period of study. Our data includes cycles with own and with donated oocytes. The total oocyte average age was 33,5±6,7. All transfers were performed at blastocyst stage. Main results and the role of chance The overall clinical pregnancy rate per embryo transfer was 70,5% (31/44), with an implantation rate of 66,7% (32/48). The mean number of transferred embryos was 1,09, with a single embryo transfer rate of 90,9%. The total mean of EPL was 0,31±0,26 [IC95% 0,23-0,39], with patients who achieved a clinical pregnancy (0,30±0,21; n = 31) showing significantly lower EPL (p-value=0,04) than the non-pregnant participants (0,41±0,40; n = 10). Three out of 44 patients had a biochemical pregnancy, with a mean EPL of 0,12±0,01. EPL cut-off value was establish at 1,1 ng/ml. The evaluation of the total dataset (n = 46 patients, including 2 outliers) showed that 8,7 % of a total of these cases had an EPL above the threshold of 1,1 ng/ml. The clinical pregnancy rate in this subgroup of patients was 25,0%. Limitations, reasons for caution This study includes a limited number of cases. Further analysis is required to confirm the present results and to allow data stratification for additional inter-group comparisons. Wider implications of the findings Evaluating EPL may contribute to improving the clinical outcome in natural modified cycles. We observed that EPL is expected to be lower in non-stimulated than in stimulated cycles, compared to previously published studies. These observations highlight the importance to manage treatment-related EPL thresholds, in particular, to establish transfer decision-making policies. Trial registration number n/a

Study question Can AI algorithms assist embryologists in evaluating embryos from any time-lapse system (TLS) along with clinical data to better predict pregnancy outcomes and reduce time-to-pregnancy? Summary answer Our algorithm (Embryoly) significantly increases accuracy in predicting clinical pregnancy by 26.9% amongst embryos deemed of fair and good quality when clinical data is included. What is known already Embryologists routinely use defined morpho-kinetic criteria to decide which embryo to transfer, and yet, many embryos deemed of good quality fail to lead to a pregnancy. Thus, AI algorithms to assist embryologists in objectively selecting the most promising embryos are in demand. To date, several reports indicate that AI algorithms are capable of predicting pregnancy clinical outcomes but to the best of our knowledge they only consider visual data (or together with a small set of clinical features) from individual TLI systems to generate their predictions. Study design, size, duration A dataset of 6790 embryos (97.82% known clinical pregnancy outcome, 31.47% frozen transfers) from 2519 patients from 11 European fertility centers recorded with 4 different TLS (GERI-Merck, Embryoscope & EmbryoscopePlus-Vitrolife and MIRI-Esco) was used to train and validate Embryoly. Nine out of 93 clinical factors were identified as being the most predictive, including woman age, woman and man BMI and AMH levels. Performances were evaluated on a separate test dataset (393 videos). Participants/materials, setting, methods Clinical pregnancy outcome was predicted using a 3D convolutional neural network that analyzed up to 5 days of embryo development. The output score was further analyzed considering the clinical features to generate a second clinical score. Both predictions were compared to those of 10 senior embryologists made on the same test dataset (with and without clinical features). Embryo quality was assessed as: poor, fair, good. Unless specified otherwise, McNemar test was used for statistical tests. Main results and the role of chance Overall accuracy of embryologists in predicting clinical pregnancy based on videos alone was 57.25% (CI 95% : 52.34% - 62.16%) compared to 60.56% (CI 95% : 55.71% - 65.41%) for Embryoly (p = 0.35). When videos were analyzed together with the clinical factors, overall accuracy of embryologists was significantly lower than Embryoly (60.05% [CI 95% : 55.19% - 64.91%] vs 68.19% [CI 95% : 63.57% - 72.82%], p-value=0.015, respectively). Clinical factors significantly increased our accuracy by 7.63% (p-value=0.030). More specifically, Embryoly algorithms fared better in terms of detecting false positives (31.30% vs 19.34%) compared to embryologists, with a specificity of 74.4% vs. 58.6%, respectively. If we consider only embryos of fair and good quality (71.50% of our test dataset) Embryoly’s accuracy was 13.52% higher than that of embryologists. This translates into AI having an even better ability to detect false positives for embryos that could be seen as good candidates for transfer (20.28% false positives against 42.70% for the embryologists). Embryoly performs differently across selected TLS when analyzing videos alone, but not when clinical data was also considered (chi2 test, p Limitations, reasons for caution As of today, Embryoly’s accuracy in predicting the outcome of poor-quality embryos is not different to that of embryologists (79.46% vs 84.96%; p-value=0.19). We are improving this by exposing Embryoly to more “poor quality” embryos, so as to also identify poor quality embryos with unexpected potential for implantation. Wider implications of the findings Our pioneering findings support the use of AI for a standardized and couple-centered care in clinical embryology, integrating male and female factors with embryo development analyses from multiple TLS. Our approach has the potential to cost-effectively reduce time to pregnancy and is another step toward a personalized embryo transfer strategy. Trial registration number Not applicable

Study question What is the SARS-CoV-2 positivity rate following the Spanish Fertility Society (SEF)/Association for the Study of Reproductive Biology (ASEBIR) screening recommendations? Summary answer The SARS-CoV-2 positivity rate in the centers following the SEF/ASEBIR screening recommendations was 0.316% after the first survey and 0.364% after the second one What is known already Due to the Sars-Cov-2 pandemic, all the Medical Assisted Reproduction (MAR) centers in Spain had to interrupt their activity most of the time during the first pandemic wave. On April 27th activity was restarted, and SEF and ASEBIR jointly elaborated a guide describing their SARS-CoV-2 screening recommendations for MAR centers. This document aims to achieve a safe environment for patients and staff. It includes the possibility of screening patients through a targeted clinical interview and the use of reverse-transcriptase polymerase chain reaction (RT-PCR). The aim of this study is to quantify the SARS-CoV-2 positivity rate based on these recommendations. Study design, size, duration National multicenter cross-sectional study. Information was gathered from centers using an anonymous survey asking for aggregated data about the number of positive cases among screened patients, sent twice. The first survey covered the period April 27th - June 30th. Second survey covered July 1st - August 31st. Response rates among centres were 9% (29/319) and 6% (20/319), respectively. This study includes 2,695 and 4,068 screenings performed in the first and the second survey, respectively. Participants/materials, setting, methods The SEF/ASEBIR recommendations describe two screening strategies. Strategy (a) consists in a targeted clinical interview (TCI) evaluating clinical symptoms and exposure risk, first before starting the cycle, and before egg-retrieval, intrauterine insemination (IUI), and/or embryo transfer (ET). Suspicious cases could be confirmed by further RT-PCR testing. Strategy (b) consists in conducting the same first TCI, and a systematic RT-PCR testing before the medical procedure in all patients. All patients in both strategies have a TCI. Main results and the role of chance In the 1st survey, 1,177 screenings and 919 RT-PCR (78.07%) were performed before the egg-retrieval. One patient with a negative TCI and positive RT-PCR was detected, and the cycle was cancelled. 1,518 screenings and 1,161 RT-PCRs (76.48%) were performed before the ET/IUI. Two patients with a positive TCI were detected, one did not perform a RT-PCR, while the other resulted in a positive RT-PCR. Both cycles were cancelled. Besides, 5 patients with negative TCI performed a RT-PCR with a positive result; all 5 were cancelled. Overall, the SARS-CoV-2 positivity rate was 8/2533 (0.316%), of which 7/2533 (0.276%) were identified by RT-PCR testing. The 2nd survey included 1,376 screenings and 1,009 RT-PCR (73.32%) performed before the egg-retrieval. Four patients with negative TCI and further positive RT-PCR were detected, and their cycle was cancelled. 2,692 screenings and 2,134 RT-PCR (79.27%) were performed before ET/IUI. Two patients had a positive TCI, one with a negative, the other with a positive RT-PCR testing; both cycles were cancelled. Besides, 8 patients with negative TCI, but positive RT-PCR testing, were detected and their cycles cancelled. Overall, the SARS-CoV-2 positivity rate was 14/3846 (0.364%), of which 13/3846 (0.338%) after positive RT-PCR testing. Limitations, reasons for caution The criteria for performing the RT-PCR testing were not the same in all MAR Centres or even in the same centre at different times. Due to the low response rate of the study, we should not extend these results to all the MAR Centres in Spain. Wider implications of the findings The results of the surveys suggest that the SEF/ASEBIR recommendations could be a good screening strategy for SARS-Cov-2 at MAR Centres. Further survey collected at different times of the pandemic are warranted, including new strategies for screening as antigen tests or vaccination status. Trial registration number Not applicable