Using deep learning to predict the outcome of live birth from more than 10,000 embryo data

Background Recently, the combination of deep learning and time-lapse imaging provides an objective, standard and scientific solution for embryo selection. However, the reported studies were based on blastocyst formation or clinical pregnancy as the end point. To the best of our knowledge, there is no predictive model that uses the outcome of live birth as the predictive end point. Can a deep learning model predict the probability of live birth from time-lapse system? Methods This study retrospectively analyzed the time-lapse data and live birth outcomes of embryos samples from January 2018 to November 2019. We used the SGD optimizer with an initial learning rate of 0.025 and cosine learning rate reduction strategy. The network is randomly initialized and trained for 200 epochs from scratch. The model is quantitively evaluated over a hold-out test and a 5-fold cross-validation by the average area under the curve (AUC) of the receiver operating characteristic (ROC) curve. Results The deep learning model was able to predict live birth outcomes from time-lapse images with an AUC of 0.968 in 5-fold stratified cross-validation. Conclusions This research reported a deep learning model that predicts the live birth outcome of a single blastocyst transfer. This efficient model for predicting the outcome of live births can automatically analyze the time-lapse images of the patient’s embryos without the need for manual embryo annotation and evaluation, and then give a live birth prediction score for each embryo, and sort the embryos by the predicted value.

A Predictive Model of Live Birth Based on Obesity and Metabolic Parameters in Patients With PCOS Undergoing Frozen-Thawed Embryo Transfer

AimsTo determine the clinical predictors of live birth in women with polycystic ovary syndrome (PCOS) undergoing frozen-thawed embryo transfer (F-ET), and to determine whether these parameters can be used to develop a clinical nomogram model capable of predicting live birth outcomes for these women.MethodsIn total, 1158 PCOS patients that were clinically pregnant following F-ET treatment were retrospectively enrolled in this study and randomly divided into the training cohort (n = 928) and the validation cohort (n = 230) at an 8:2 ratio. Relevant risk factors were selected via a logistic regression analysis approach based on the data from patients in the training cohort, and odds ratios (ORs) were calculated. A nomogram was constructed based on relevant risk factors, and its performance was assessed based on its calibration and discriminative ability.ResultsIn total, 20 variables were analyzed in the present study, of which five were found to be independently associated with the odds of live birth in univariate and multivariate logistic regression analyses, including advanced age, obesity, total cholesterol (TC), triglycerides (TG), and insulin resistance (IR). Having advanced age (OR:0.499, 95% confidence interval [CI]: 0.257 – 967), being obese (OR:0.506, 95% CI: 0.306 - 0.837), having higher TC levels (OR: 0.528, 95% CI: 0.423 - 0.660), having higher TG levels (OR: 0.585, 95% CI: 0.465 - 737), and exhibiting IR (OR:0.611, 95% CI: 0.416 - 0.896) were all independently associated with a reduced chance of achieving a live birth. A predictive nomogram incorporating these five variables was found to be well-calibrated and to exhibit good discriminatory capabilities, with an area under the curve (AUC) for the training group of 0.750 (95% CI, 0.709 - 0.788). In the independent validation cohort, this model also exhibited satisfactory goodness-of-fit and discriminative capabilities, with an AUC of 0.708 (95% CI, 0.615 - 0.781).ConclusionsThe nomogram developed in this study may be of value as a tool for predicting the odds of live birth for PCOS patients undergoing F-ET, and has the potential to improve the efficiency of pre-transfer management.

The effect of cult-active medium on pregnancy outcomes after intracytoplasmic sperm injection in azoospermic men: A case-control study

Background: Failed oocyte activation following intracytoplasmic sperm injection (ICSI) as a result of calcium deficiency is a major challenge. Objective: We compared the effect of cult-active medium (CAM) on ICSI outcomes in obstructive azoospermia cases. Materials and Methods: The present study was conducted with 152 ICSI cases, classified into CAM and control groups. The injected oocytes in the control group were cultured in the cleavage medium, while in the artificial oocyte activation group, oocytes were chemically activated through exposure to 200 µL of CAM for 15 min. Fertilization and cleavage rates, quality of embryos, and biochemical pregnancy and live birth rates were assessed in both groups. Results: There were significant differences between the groups in terms of fertilization and cleavage rates after using the CAM in the percutaneous epididymal sperm aspiration (PESA) subgroup (p = 0.05, p ≤ 0.001) and in the testicular sperm extraction subgroup (p = 0.02, p = 0.04), compared to their control groups. Also, the pregnancy rate was significantly higher in the PESA-CAM subgroup (p = 0.03). The PESA-CAM subgroup demonstrated a significant difference in embryo quality after ICSI (p = 0.04). Unsuccessful embryo transfer and abortion were lower in both subgroups compared to the control groups, but this difference was not significant. Surprisingly, live birth rate was higher in the PESA-CAM subgroup (p = 0.03). Conclusion: CAM treatment could improve fertilization and cleavage rates in obstructive azoospermia participants. It had a significant effect on embryo quality, and pregnancy and live birth rates in PESA cases. Key words: Calcium ionophore, Obstructive azoospermia, Fertilization, ICSI.

Comparison of ICSI and Conventional IVF in Non-Male Factor Patients with Less Than Four Oocytes

Background: We aimed to analyse our clinical results for a particular subgroup of patients with poor ovarian response (POR) in order to clarify if lower number of oocytes is a drawback for proceeding to C-IVFMaterials and methods: In this retrospective study, patient files of all couples (#1733) who underwent oocyte retrieval between January 2017 and December 2019 were reviewed and 191 cases diagnosed with non-male factor infertility in which ≤3 cumulus-oocyte-complexes available for fertilisation were analysed. Exclusion criteria were: woman age>42, patients with a history of previous ART trial, prenatal genetic testing cycles and couples undergoing total cryopreservation for any indication. Three groups were constructed depending on the method of fertilisation and on semen quality as follows: IVF non-male factor (Group 1,n=77); ICSI non-male factor (Group 2, n=65); ICSI male factor- ICSI/MF n=49 according to WHO reference values. Main outcome parameters were: fertilisation rate, implantation rate and live birth rate. Results: Fertilisation rate per collected COC was significantly higher in group 1 compared to the other two groups (85,68%, 72,58%, 73,33% respectively, p=0,004). FR per inseminated oocyte also tended to be higher in group 1 but not reaching a statistically significant level. Both techniques yielded similar implantation rates (20,42%, 28,49%, 23,33% respectively, p=0,407) and live birth rates (26,8%, 30,6%, 31,1% respectively, p=0.643).Conclusions: In the presence of normal semen parameters, low egg number is not an indication to perform ICSI. The choice of fertilisation method should be based primarily on semen quality, in combination with the patient’s previous history regardless of the ovarian reserve.

Oocyte cryopreservation review: outcomes of medical oocyte cryopreservation and planned oocyte cryopreservation

Background The utilization of oocyte cryopreservation (OC) has become popularized with increasing numbers of reproductive-aged patients desiring to maintain fertility for future family building. OC was initially used for fertility preservation in postmenarchal patients prior to gonadotoxic therapies; however, it is now available to patients to circumvent age-related infertility and other diagnoses associated with early loss of ovarian reserve. The primary aim of this paper is to provide a narrative review of the most recent and robust data on the utilization and outcomes of OC in both patient populations. Summary OC results in similar oocyte yield in patients facing gonadotoxic therapies and patients undergoing planned OC. Available data are insufficient to predict the live birth rates or the number of oocytes needed to result in live birth. However, oocyte yield and live birth rates are best among patients < 37.5 years old or with anti-mullerian hormone levels > 1.995 ng/dL, at the time of oocyte retrieval. There is a high ‘no use’ rate (58.9%) in patients using planned OC with 62.5% returning to use frozen oocytes with a spouse. The utilization rate in medical OC patients is < 10%. There is currently no data on the effects of BMI, smoking, or ethnicity on planned OC outcomes. Conclusion It is too early to draw any final conclusions on outcomes of OC in medical OC and planned OC; however, preliminary data supports that utilization of OC in both groups result in preservation of fertility and subsequent live births in patients who return to use their cryopreserved eggs. Higher oocyte yield, with fewer ovarian stimulation cycles, and higher live birth rates are seen in patients who seek OC at younger ages, reinforcing the importance of age on fertility preservation. More studies are needed in medical OC and planned OC to help guide counseling and decision-making in patients seeking these services.

Endometrial microbiota composition is associated with reproductive outcome in infertile patients

Background Previous evidence indicates associations between the female reproductive tract microbiome composition and reproductive outcome in infertile patients undergoing assisted reproduction. We aimed to determine whether the endometrial microbiota composition is associated with reproductive outcomes of live birth, biochemical pregnancy, clinical miscarriage or no pregnancy. Methods Here, we present a multicentre prospective observational study using 16S rRNA gene sequencing to analyse endometrial fluid and biopsy samples before embryo transfer in a cohort of 342 infertile patients asymptomatic for infection undergoing assisted reproductive treatments. Results A dysbiotic endometrial microbiota profile composed of Atopobium, Bifidobacterium, Chryseobacterium, Gardnerella, Haemophilus, Klebsiella, Neisseria, Staphylococcus and Streptococcus was associated with unsuccessful outcomes. In contrast, Lactobacillus was consistently enriched in patients with live birth outcomes. Conclusions Our findings indicate that endometrial microbiota composition before embryo transfer is a useful biomarker to predict reproductive outcome, offering an opportunity to further improve diagnosis and treatment strategies.

Cumulative Live Birth Rates for Low-Prognosis Women Over 5 Years or 9 Frozen-Thawed Embryo Transfer Cycles

Background: For heterogeneous populations of low-prognosis women, it remains unclear as to how long individuals should continue undergoing ART when attempting to have a baby, as there have been insufficient studies to date tracking the cumulative live birth rates (CLBRs) for these women over the entire course of their ART treatment, particularly over extended time periods. Methods: This was a retrospective analysis of 17,698 women at a tertiary care academic medical center who had begun undergoing IVI/ICSI cycles using a PPOS approach between January 2013 and January 2019. Low-prognosis patients were stratified into four groups based upon POSEIDON criteria, with patients exhibiting normal or high ovarian reserves and response to stimulation (defined as AFC ≥5, >9 oocytes retrieved) being included as controls (group 5). The CLBR within 5 years or 9 FET cycles from the OPU day of the first cycle was the primary endpoint for this study, including all repetitive oocyte retrieval cycles and subsequent FET cycles. Optimistic and conservative approaches were used for the analysis of CLBRs and the depiction of cumulative incidence curves.Results: Under both optimistic and conservative model analyses, patients in group 5 exhibited the highest CLBR within 5 years or 9 FET cycles, followed by POSEIDON group 1, group 3, group 2, and group 4. Upward trends in CLBRs were evident across the five groups with the prolongation of time or an increase in FET cycle counts. Within the first 2 years or 3 FET cycles, the CLBRs rose rapidly, followed by more moderate increases over the following 2-3.5 years or 4-6 cycles, with the patients in group 3 and group 4 exhibiting the most obvious improvements. Patients in all groups reached a CLBR plateau after 3.5 years or 6 FET cycles.Conclusions: All low-prognosis women should undergo ART treatment for a minimum of 2 years or 3 FET cycles, and exhibit better outcomes when extending ART treatment to 3.5 years or 6 FET cycles (particularly for POSEIDON groups 3 and 4), but should consider ceasing further treatment thereafter due to a lack of apparent benefit.