Predicting the number of oocytes retrieved from controlled ovarian hyperstimulation with machine learning.
Ontology highlight
ABSTRACT:
SUBMITTER: Ferrand T
PROVIDER: S-EPMC10546073 | biostudies-literature | 2023 Oct
REPOSITORIES: biostudies-literature
Publications
Predicting the number of oocytes retrieved from controlled ovarian hyperstimulation with machine learning.
Ferrand Timothy T Boulant Justine J He Chloe C Chambost Jérôme J Jacques Céline C Pena Chris-Alexandre CA Hickman Cristina C Reignier Arnaud A Fréour Thomas T
Human reproduction (Oxford, England) 20231001 10
<h4>Study question</h4>Can machine learning predict the number of oocytes retrieved from controlled ovarian hyperstimulation (COH)?<h4>Summary answer</h4>Three machine-learning models were successfully trained to predict the number of oocytes retrieved from COH.<h4>What is known already</h4>A number of previous studies have identified and built predictive models on factors that influence the number of oocytes retrieved during COH. Many of these studies are, however, limited in the fact that they ...[more]
Similar Datasets
Project description:ObjectiveOvarian reserve tests are commonly used to predict ovarian response in infertile patients undergoing ovarian stimulation. Although serum markers such as basal follicle-stimulating hormone (FSH) or random anti-Müllerian hormone (AMH) level and ultrasonographic markers (antral follicle count, AFC) are good predictors, no single test has proven to be the best predictor. In this study, we developed appropriate equations and novel nomograms to predict the number of oocytes that will be retrieved using patients' age, serum levels of basal FSH and AMH, and AFC.MethodsWe analyzed a database containing clinical and laboratory information of 141 stimulated in vitro fertilization (IVF) cycles performed at a university-based hospital between September 2009 and December 2013. We used generalized linear models for prediction of the number of oocytes.ResultsAge, basal serum FSH level, serum AMH level, and AFC were significantly related to the number of oocytes retrieved according to the univariate and multivariate analyses. The equations that predicted the number of oocytes retrieved (log scale) were as follows: model (1) 3.21-0.036×(age)+0.089×(AMH), model (2) 3.422-0.03×(age)-0.049×(FSH)+0.08×(AMH), model (3) 2.32-0.017×(age)+0.039×(AMH)+0. 03×(AFC), model (4) 2.584-0.015×(age)-0.035×(FSH)+0.038×(AMH)+0.026×(AFC). model 4 showed the best performance. On the basis of these variables, we developed nomograms to predict the number of oocytes that can be retrieved.ConclusionOur nomograms helped predict the number of oocytes retrieved in stimulated IVF cycles.
| S-EPMC4925866 | biostudies-literature
Project description:BackgroundPredicting the number of oocytes retrieved (NOR) following controlled ovarian stimulation (COS) is the only way to ensure effective and safe treatment in assisted reproductive technology (ART). To date, there have been limited studies about predicting specific NOR, which hinders the development of individualized treatment in ART.ObjectiveTo establish an online tool for predicting NOR.Materials and methodsIn total, 621 prospective routine gonadotropin releasing hormone (GnRH) antagonist COS cycles were studied. Independent variables included age, body mass index, antral follicle counts, basal FSH, basal and increment of anti-mullerian hormone, Luteinizing hormon, estradiol, testosterone, androstenedione, and inhibin B. The outcome variable was NOR. The independent variables underwent appropriate transformation to achieve a better fit for a linear relationship with NOR. Pruned forward selection with holdback validation was then used to establish predictive models. Corrected Akaike's information criterion, Schwarz-Bayesian information criterion, scaled -log[likelihood], and the generalized coefficient of determination (R 2) were used for model evaluation.ResultsA multiple negative binomial regression model was used for predicting NOR because it fitted a negative binomial distribution. We established Model 1, using basal ovarian reserve markers, and Model 2, using both basal and early dynamic markers for predicting NOR following COS. The generalized R 2 values were 0.54 and 0.51 for Model 1 and 0.64 and 0.62 for Model 2 in the training and validation sets, respectively.ConclusionModels 1 and 2 could be applied to different scenarios. For directing the starting dose of recombinant follicle stimulation hormone (rFSH), Model 1 using basic predictors could be used prior to COS. Model 2 could be used for directing the adjustment of rFSH dosages during COS. An online tool (http://121.43.113.123:8002/) based on these two models is also developed. We anticipate that the clinical application of this tool could help the ART clinics to reduce iatrogenic ovarian under- or over-responses, and could reduce costs during COS for ART.
| S-EPMC9186016 | biostudies-literature
Project description:PurposeTo investigate whether the number of oocytes retrieved after controlled ovarian stimulation (COS) for fertility preservation (FP) can be predicted using serum anti-Müllerian hormone (AMH) levels in adolescent and young adult (AYA) women.MethodsThis is a retrospective cohort study in a large university-affiliated fertility center. AYA women aged 11 to 25 years received COS using gonadotropin-releasing hormone (GnRH) antagonist protocols for FP were enrolled. Those with canceled cycles or incomplete data were excluded. The primary outcome is to determine whether the number of oocytes retrieved can be predicted through serum AMH levels using multiple linear regression analysis.ResultsThe mean numbers of oocytes retrieved, mature oocytes retrieved, and oocytes cryopreserved were 10.3 ± 7.6, 6.9 ± 6.2, and 8.5 ± 6.8, respectively. Multiple stepwise linear regression analysis revealed that serum AMH level independently predicts COS outcomes, including the total number of oocytes retrieved (adjusted R2 = 0.222, P < 0.001), the number of mature oocytes retrieved (adjusted R2 = 0.102, P = 0.013), and the number of oocytes cryopreserved (adjusted R2 = 0.153, P = 0.003).ConclusionSerum AMH level was a significant predictor of the number of oocytes retrieved after COS for FP in AYA women, while antral follicle count and age were not significantly related.
| S-EPMC11920294 | biostudies-literature
Project description:Controlled ovarian stimulation is tailored to the patient based on clinical parameters but estimating the number of retrieved metaphase II (MII) oocytes is a challenge. Here, we have developed a model that takes advantage of the patient's genetic and clinical characteristics simultaneously for predicting the stimulation outcome. Sequence variants in reproduction-related genes identified by next-generation sequencing were matched to groups of various MII oocyte counts using ranking, correspondence analysis, and self-organizing map methods. The gradient boosting machine technique was used to train models on a clinical dataset of 8,574 or a clinical-genetic dataset of 516 ovarian stimulations. The clinical-genetic model predicted the number of MII oocytes better than that based on clinical data. Anti-Müllerian hormone level and antral follicle count were the two most important predictors while a genetic feature consisting of sequence variants in the GDF9, LHCGR, FSHB, ESR1, and ESR2 genes was the third. The combined contribution of genetic features important for the prediction was over one-third of that revealed for anti-Müllerian hormone. Predictions of our clinical-genetic model accurately matched individuals' actual outcomes preventing over- or underestimation. The genetic data upgrades the personalized prediction of ovarian stimulation outcomes, thus improving the in vitro fertilization procedure.
| S-EPMC10138216 | biostudies-literature
Project description:The objective of this study was to investigate the differences in the gene expression profile of the granulosa cells from preovulatory follicles obtained after controlled ovarian hyperstimulation (COH) with either recombinant FSH (rFSH) or urine derived human menopausal gonadotropins (hMG).
2008-01-31 | E-MEXP-1238 | biostudies-arrayexpress
Project description:BackgroundThe aim of this study was to determine the relationship between a purported luteinizing hormone/chorionic gonadotropin (LHCGR) high function polymorphism (rs4539842/insLQ) and outcome to controlled ovarian hyperstimulation (COH).MethodsThis was a prospective study of 172 patients undergoing COH at the Fertility and IVF Center at GWU. DNA was isolated from blood samples and a region encompassing the insLQ polymorphism was sequenced. We also investigated a polymorphism (rs4073366 G > C) that was 142 bp from insLQ. The association of the insLQ and rs4073366 alleles and outcome to COH (number of mature follicles, estradiol level on day of human chorionic gonadotropin (hCG) administration, the number of eggs retrieved and ovarian hyperstimulation syndrome (OHSS)) was determined.ResultsIncreasing age and higher day 3 (basal) FSH levels were significantly associated with poorer response to COH. We found that both insLQ and rs4073366 were in linkage disequilibrium (LD) and no patients were homozygous for both recessive alleles (insLQ/insLQ; C/C). The insLQ variant was not significantly associated with any of the main outcomes to COH. Carrier status for the rs4073366 C variant was associated (P = 0.033) with an increased risk (OR 2.95, 95% CI = 1.09-7.96) of developing OHSS.ConclusionsWhile age and day 3 FSH levels were predictive of outcome, we found no association between insLQ and patient response to COH. Interestingly, rs4073366 C variant carrier status was associated with OHSS risk. To the best of our knowledge, this is the first report suggesting that LHCGR genetic variation might function in patient risk for OHSS.
| S-EPMC3727944 | biostudies-literature
Project description:BackgroundFertility preservation (FP) protocols in case of breast cancer (BC) include mature oocyte cryopreservation following letrozole associated controlled ovarian hyperstimulation (Let-COH). To date, the impact of Let-COH on the follicular microenvironment has been poorly investigated, although a high androgen/estrogen ratio was previously associated with low oocyte quality.MethodsIn this prospective study, follicular fluid (FF) steroid levels (estradiol, testosterone, progesterone) and cumulus cell (CC) gene expression related to oocyte quality (HAS2, PTGS2, GREM1) were compared between 23 BC patients undergoing Let-COH for FP and 24 infertile patients undergoing conventional COH without letrozole. All patients underwent an antagonist COH cycle, and ovulation was triggered with hCG or GnRHa in both groups.ResultsFF estradiol levels were significantly lower while testosterone levels were significantly higher in the study group compared to controls irrespective of the trigger method. However, estradiol levels increased significantly with GnRHa triggering compared to hCG in the study group (median = 194.5 (95.4-438) vs 64.4 (43.8-152.4) ng/ml, respectively, p < 0.001), but not in the control group (median = 335.5 (177.5-466.7) vs 354 (179-511) ng/ml, respectively). After hCG trigger, Cumulus cell (CC) gene expression was lower in the study group compared to the control group, and difference was significant for PTGS2. Conversely, CC gene expression of PTGS2 and GREM1 was significantly higher in the study group compared to controls when ovulation was triggered with GnRHa.ConclusionsLet-COH triggered with hCG may negatively impact oocyte quality. However, ovulation triggering with GnRHa may improve the oocyte microenvironment and cumulus cell genes expression in Let-COH, suggesting a positive impact on oocyte quality in breast cancer patients.Trial registrationClinicaltrials.gov - NCT02661932 , registered 25 January 2016, retrospectively registered.
| S-EPMC6318989 | biostudies-literature
Project description:ObjectiveThe study looked into the possible influence of GDF9 polymorphisms on ovarian response in women with a normal ovarian reserve undergoing controlled ovarian hyperstimulation for in vitro fertilization (IVF).MethodsThis cross-sectional study included 67 women with normal ovarian reserve aged 30-39 years submitted to controlled ovarian hyperstimulation for IVF. We sequenced four polymorphisms in the GDF9 gene (C398G, C447T, G546A, and G646A) and analyzed their influence on follicular and oocyte outcomes.ResultsThe mutant allele C398G decreased the total number of follicles >17mm (6.49 vs. 4.33, p=0.001), total number of follicles (10.11 vs. 7.33, p=0.032), number of MII oocytes retrieved, and serum progesterone levels on trigger day. The C447T polymorphism was associated with a greater number of follicles between 12 and 14 mm on the day of r-hCG, while the G546A polymorphism was associated with lower serum progesterone levels on trigger day.ConclusionsGDF9 gene polymorphisms C398G and C447T adversely affected ovarian response in women undergoing controlled ovarian hyperstimulation. These findings show that in addition to playing a role in the early stages of folliculogenesis, GDF9 polymorphisms have an important impact on the final stage of oocyte development.
| S-EPMC7558902 | biostudies-literature
Project description:BackgroundThe association of recombinant FSH (rFSH) plus recombinant LH (rLH) is currently used for Controlled Ovarian Stimulation (COS) in human IVF, but its efficacy has, to date, not yet been compared to that of human Menopausal Gonadotropin (hMG), the FSH + LH activity-containing urinary drug.MethodsEight hundred forty-eight (848) IVF patients classified as expected "poor" or "normal" responders according to antral follicle count (AFC) and basal (day 3) FSH were treated with rFSH + rLH (2:1 ratio, n = 398, Group A) or hMG (n = 450, Group B). Data were collected under real-life practice circumstances and the pregnancy rate with fresh embryos was calculated by stratifying patients according to the number of retrieved oocytes (1-2, 3-4, 5-6, 7-8, >8).ResultsOverall, the pregnancy rate in both groups progressively improved according to the number of oocytes retrieved. When comparing patients within the same subgroup of oocyte yield, Group A and B showed a comparable outcome up to the reported highest yield (>8). When more than 8 oocytes were available, Group A had a significantly better pregnancy rate outcome. Patients' characteristics did not significantly differ between the two groups and the better outcome in the best responding patients in Group A was confirmed by a multivariable logistic regression analysis, that showed that both the use of rFSH + rLH and the total number of retrieved oocytes increased the probability of pregnancy with odd ratio (OR) of 1.628 and 1.083, respectively.ConclusionsWhen comparing patients with the same number of retrieved oocytes under real-life circumstances, the association of rFSH + rLH results in a significantly higher pregnancy rate than hMG when more than 8 oocytes are retrieved. The reason(s) for this are unknown, but a more favorable effect on oocyte quality and/or endometrial receptivity could be involved.
| S-EPMC4514947 | biostudies-literature
Project description:PurposeTo assess whether preimplantation genetic testing for aneuploidies (PGT-A) at the blastocyst stage improves clinical outcomes compared with transfer of embryos without PGT-A in poor ovarian response (POR) patients.MethodsRetrospective cohort study of IVF cycles from 2016 to 2019 at a single academic fertility center. IVF cycles with POR and four or fewer oocytes retrieved were stratified into PGT-A (n = 241) and non-PGT (n = 112) groups. In PGT-A cycles, trophectoderm biopsy, next-generation sequencing with 24-chromosome screening, and single euploid frozen embryo transfer were performed. In non-PGT cycles, fresh or frozen transfer of untested embryos on day 3 or 5 was performed. Main outcomes included live birth rate and miscarriage rate per retrieval.Result(s)Patients who underwent PGT-A cycles were significantly less likely to reach embryo transfer compared with those who underwent non-PGT cycles (13.7% vs 70.6%). The live birth rate per retrieval did not differ between the PGT-A and non-PGT groups (6.6% vs 5.4%). Overall, the miscarriage rate was low. The PGT-A group demonstrated a significantly lower miscarriage rate per retrieval (0.4% vs 3.6%) as well as per pregnancy (5.9% vs 40.0%) compared with the non-PGT group. The number needed to treat to avoid one clinical miscarriage was 31 PGT-A cycles.Conclusion(s)PGT-A did not improve live birth rate per retrieval in POR patients with four or fewer oocytes retrieved. PGT-A was associated with a lower miscarriage rate; however, a fairly large number of PGT-A cycles were needed to prevent one miscarriage.
| S-EPMC7244670 | biostudies-literature