Evaluation of artificial intelligence using time-lapse images of IVF embryos to predict live birth

Clinical prediction models for in vitro fertilization outcomes: a systematic review, meta-analysis, and external validation

STUDY QUESTION

What is the best-performing model currently predicting live birth outcomes for IVF or ICSI?

SUMMARY ANSWER

Among the identified prognostic models, McLernon's post-treatment model outperforms other models in both the meta-analysis and external validation of a Chinese cohort.

WHAT IS KNOWN ALREADY

With numerous similar models available across different time periods and using various predictors in IVF prognostic models, there is a need to summarize and evaluate them, due to a lack of validated evidence distinguishing high-quality from low-quality prediction tools. However, there is a notable dearth of research in the form of meta-analysis or external validation assessing the performance of models in predicting live births in this field.

STUDY DESIGN, SIZE, DURATION

The researchers conducted a comprehensive literature review in PubMed, EMBASE, and Web of Science, using keywords related to prognostic models and IVF/ICSI live birth outcomes. The search included studies published up to 3 April 2024, and was limited to English language studies.

PARTICIPANTS/MATERIALS, SETTING, METHODS

The review included studies that developed or validated prognostic models for IVF live birth outcomes while providing clear reports on model characteristics. Researchers extracted and analysed the data in accordance with the guidelines outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses and other model-related guidelines. For model effects in meta-analysis, the choice would be based on the heterogeneity assessed using the I2 statistic and the Cochrane Q test. Model performance was evaluated by assessing their area under the receiver operating characteristic curves (AUCs) and calibration plots in the studies.

MAIN RESULTS AND THE ROLE OF CHANCE

This review provides a comprehensive summary of data derived from 72 studies with an overall ROB of high or unclear. These studies contained a total of 132 predictors and 86 prognostic models, and then meta-analyses were performed for each of the five selected models. The total random effects of Templeton's, Nelson's, McLernon's pre-treatment and post-treatment model demonstrated AUCs of 0.65 (95% CI: 0.61-0.69), 0.63 (95% CI: 0.63-0.64), 0.67 (95% CI: 0.62-0.71), and 0.73 (95% CI: 0.71-0.75), respectively. The total fixed effects of the intelligent data analysis score (iDAScore) model estimated an AUC of 0.66 (95% CI: 0.63-0.68). The external validation of the initial four models in our cohort produced AUCs ranging from 0.53 to 0.58, and the calibration was confirmed through calibration plots.

LIMITATIONS, REASONS FOR CAUTION

While the focus on English-language studies and live birth outcomes may constrain the generalizability of the findings to diverse populations, this approach equips clinicians, who view live births as the ultimate objective, with more precise and actionable reference guidelines.

WIDER IMPLICATIONS OF THE FINDINGS

This study represents the first meta-analysis in the field of IVF prediction models, definitively confirming the superior performance of McLernon's post-treatment model. The conclusion is reinforced by independent validation from another perspective. Nevertheless, further investigation is warranted to develop new models and to externally validate existing high-performing models for prognostic accuracy in IVF outcomes.

STUDY FUNDING/COMPETING INTEREST(S)

This study was supported by the National Natural Science Foundation of China (Grant No. 82174517). The authors report no conflict of interest.

REGISTRATION NUMBER

2022 CRD42022312018.

Catalyzing IVF outcome prediction: exploring advanced machine learning paradigms for enhanced success rate prognostication

This study addresses the research problem of enhancing In-Vitro Fertilization (IVF) success rate prediction by integrating advanced machine learning paradigms with gynecological expertise. The methodology involves the analysis of comprehensive datasets from 2017 to 2018 and 2010-2016. Machine learning models, including Logistic Regression, Gaussian NB, SVM, MLP, KNN, and ensemble models like Random Forest, AdaBoost, Logit Boost, RUS Boost, and RSM, were employed. Key findings reveal the significance of patient demographics, infertility factors, and treatment protocols in IVF success prediction. Notably, ensemble learning methods demonstrated high accuracy, with Logit Boost achieving an accuracy of 96.35%. The implications of this research span clinical decision support, patient counseling, and data preprocessing techniques, highlighting the potential for personalized IVF treatments and continuous monitoring. The study underscores the importance of collaboration between gynecologists and data scientists to optimize IVF outcomes. Prospective studies and external validation are suggested as future directions, promising to further revolutionize fertility treatments and offer hope to couples facing infertility challenges.

Deep learning versus manual morphology-based embryo selection in IVF: a randomized, double-blind noninferiority trial

To assess the value of deep learning in selecting the optimal embryo for in vitro fertilization, a multicenter, randomized, double-blind, noninferiority parallel-group trial was conducted across 14 in vitro fertilization clinics in Australia and Europe. Women under 42 years of age with at least two early-stage blastocysts on day 5 were randomized to either the control arm, using standard morphological assessment, or the study arm, employing a deep learning algorithm, intelligent Data Analysis Score (iDAScore), for embryo selection. The primary endpoint was a clinical pregnancy rate with a noninferiority margin of 5%. The trial included 1,066 patients (533 in the iDAScore group and 533 in the morphology group). The iDAScore group exhibited a clinical pregnancy rate of 46.5% (248 of 533 patients), compared to 48.2% (257 of 533 patients) in the morphology arm (risk difference -1.7%; 95% confidence interval -7.7, 4.3; P = 0.62). This study was not able to demonstrate noninferiority of deep learning for clinical pregnancy rate when compared to standard morphology and a predefined prioritization scheme. Australian New Zealand Clinical Trials Registry (ANZCTR) registration: 379161 .

Identifying key predictive features for live birth rate in advanced maternal age patients undergoing single vitrified-warmed blastocyst transfer

BACKGROUND

Infertility affects one in six couples worldwide, with advanced maternal age (AMA) posing unique challenges due to diminished ovarian reserve and reduced oocyte quality. Single vitrified-warmed blastocyst transfer (SVBT) has shown promise in assisted reproductive technology (ART), but success rates in AMA patients remain suboptimal. This study aimed to identify and refine predictive factors for live birth following SVBT in AMA patients, with the goal of enhancing clinical decision-making and enabling personalized treatment strategies.

METHODS

This retrospective cohort study analyzed 1,168 SVBT cycles conducted between June 2016 and December 2022 at the First Affiliated Hospital of Guangxi Medical University and Nanning Maternity and Child Health Hospital. Nineteen machine-learning models were applied to identify key predictive factors for live birth. Feature selection and 10-fold cross-validation were employed to validate the models.

RESULTS

The most significant predictors of live birth included inner cell mass quality, trophectoderm quality, number of oocytes retrieved, endometrial thickness, and the presence of 8-cell blastomeres on day 3. The stacking model demonstrated the best predictive performance (AUC: 0.791), followed by Extra Trees (AUC: 0.784) and Random Forest (AUC: 0.768). These models outperformed traditional methods, achieving superior accuracy, sensitivity, and specificity.

CONCLUSION

Leveraging advanced machine-learning models and identifying critical predictive factors can improve the accuracy of live birth outcome predictions for AMA patients undergoing SVBT. These findings offer valuable insights for enhancing clinical decision-making and managing patient expectations. Further research is needed to validate these results in larger, multi-center cohorts and to explore additional factors, including fresh embryo transfers, to broaden the applicability of these models in clinical practice.

Visual interpretability of image-based classification models by generative latent space disentanglement applied to in vitro fertilization

The success of deep learning in identifying complex patterns exceeding human intuition comes at the cost of interpretability. Non-linear entanglement of image features makes deep learning a "black box" lacking human meaningful explanations for the models' decision. We present DISCOVER, a generative model designed to discover the underlying visual properties driving image-based classification models. DISCOVER learns disentangled latent representations, where each latent feature encodes a unique classification-driving visual property. This design enables "human-in-the-loop" interpretation by generating disentangled exaggerated counterfactual explanations. We apply DISCOVER to interpret classification of in vitro fertilization embryo morphology quality. We quantitatively and systematically confirm the interpretation of known embryo properties, discover properties without previous explicit measurements, and quantitatively determine and empirically verify the classification decision of specific embryo instances. We show that DISCOVER provides human-interpretable understanding of "black box" classification models, proposes hypotheses to decipher underlying biomedical mechanisms, and provides transparency for the classification of individual predictions.

Deep learning pipeline reveals key moments in human embryonic development predictive of live birth after in vitro fertilization

Demand for in vitro fertilization (IVF) treatment is growing; however, success rates remain low partly due to difficulty in selecting the best embryo to be transferred. Current manual assessments are subjective and may not take advantage of the most informative moments in embryo development. Here, we apply convolutional neural networks (CNNs) to identify key windows in pre-implantation human development that can be linked to embryo viability and are therefore suitable for the early grading of IVF embryos. We show how machine learning models trained at these developmental time points can be used to refine overall embryo viability assessment. Exploiting the well-known capabilities of transfer learning, we illustrate the performance of CNN models for very limited datasets, paving the way for the use on a clinic-by-clinic basis, catering for local data heterogeneity.

Viewing early life without labels: optical approaches for imaging the early embryo†

Embryo quality is an important determinant of successful implantation and a resultant live birth. Current clinical approaches for evaluating embryo quality rely on subjective morphology assessments or an invasive biopsy for genetic testing. However, both approaches can be inherently inaccurate and crucially, fail to improve the live birth rate following the transfer of in vitro produced embryos. Optical imaging offers a potential non-invasive and accurate avenue for assessing embryo viability. Recent advances in various label-free optical imaging approaches have garnered increased interest in the field of reproductive biology due to their ability to rapidly capture images at high resolution, delivering both morphological and molecular information. This burgeoning field holds immense potential for further development, with profound implications for clinical translation. Here, our review aims to: (1) describe the principles of various imaging systems, distinguishing between approaches that capture morphological and molecular information, (2) highlight the recent application of these technologies in the field of reproductive biology, and (3) assess their respective merits and limitations concerning the capacity to evaluate embryo quality. Additionally, the review summarizes challenges in the translation of optical imaging systems into routine clinical practice, providing recommendations for their future development. Finally, we identify suitable imaging approaches for interrogating the mechanisms underpinning successful embryo development.

An artificial intelligence algorithm to select most viable embryos considering current process in IVF labs

Background

The most common Assisted Reproductive Technology is In-Vitro Fertilization (IVF). During IVF, embryologists commonly perform a morphological assessment to evaluate embryo quality and choose the best embryo for transferring to the uterus. However, embryo selection through morphological assessment is subjective, so various embryologists obtain different conclusions. Furthermore, humans can consider only a limited number of visual parameters resulting in a poor IVF success rate. Artificial intelligence (AI) for embryo selection is objective and can include many parameters, leading to better IVF outcomes.

Objectives

This study sought to use AI to (1) predict pregnancy results based on embryo images, (2) assess using more than one image of the embryo in the prediction of pregnancy but based on the current process in IVF labs, and (3) compare results of AI-Based methods and embryologist experts in predicting pregnancy.

Methods

A data set including 252 Time-lapse Videos of embryos related to IVF performed between 2017 and 2020 was collected. Frames related to 19 ± 1, 43 ± 1, and 67 ± 1 h post-insemination were extracted. Well-Known CNN architectures with transfer learning have been applied to these images. The results have been compared with an algorithm that only uses the final image of embryos. Furthermore, the results have been compared with five experienced embryologists.

Results

To predict the pregnancy outcome, we applied five well-known CNN architectures (AlexNet, ResNet18, ResNet34, Inception V3, and DenseNet121). DeepEmbryo, using three images, predicts pregnancy better than the algorithm that only uses one final image. It also can predict pregnancy better than all embryologists. Different well-known architectures can successfully predict pregnancy chances with up to 75.0% accuracy using Transfer Learning.

Conclusion

We have developed DeepEmbryo, an AI-based tool that uses three static images to predict pregnancy. Additionally, DeepEmbryo uses images that can be obtained in the current IVF process in almost all IVF labs. AI-based tools have great potential for predicting pregnancy and can be used as a proper tool in the future.

Diagnostic or prognostic? Decoding the role of embryo selection on in vitro fertilization treatment outcomes

In this review, we take a fresh look at embryo assessment and selection methods from the perspective of diagnosis and prognosis. On the basis of a systematic search in the literature, we examined the evidence on the prognostic value of different embryo assessment methods, including morphological assessment, blastocyst culture, time-lapse imaging, artificial intelligence, and preimplantation genetic testing for aneuploidy.

Associations between the artificial intelligence scoring system and live birth outcomes in preimplantation genetic testing for aneuploidy cycles

BACKGROUND

Several studies have demonstrated that iDAScore is more accurate in predicting pregnancy outcomes in cycles without preimplantation genetic testing for aneuploidy (PGT-A) compared to KIDScore and the Gardner criteria. However, the effectiveness of iDAScore in cycles with PGT-A has not been thoroughly investigated. Therefore, this study aims to assess the association between artificial intelligence (AI)-based iDAScore (version 1.0) and pregnancy outcomes in single-embryo transfer (SET) cycles with PGT-A.

METHODS

This retrospective study was approved by the Institutional Review Board of Chung Sun Medical University, Taichung, Taiwan. Patients undergoing SET cycles (n = 482) following PGT-A at a single reproductive center between January 2017 and June 2021. The blastocyst morphology and morphokinetics of all embryos were evaluated using a time-lapse system. The blastocysts were ranked based on the scores generated by iDAScore, which were defined as AI scores, or by KIDScore D5 (version 3.2) following the manufacturer's protocols. A single blastocyst without aneuploidy was transferred after examining the embryonic ploidy status using a next-generation sequencing-based PGT-A platform. Logistic regression analysis with generalized estimating equations was conducted to assess whether AI scores are associated with the probability of live birth (LB) while considering confounding factors.

RESULTS

Logistic regression analysis revealed that AI score was significantly associated with LB probability (adjusted odds ratio [OR] = 2.037, 95% confidence interval [CI]: 1.632-2.542) when pulsatility index (PI) level and types of chromosomal abnormalities were controlled. Blastocysts were divided into quartiles in accordance with their AI score (group 1: 3.0-7.8; group 2: 7.9-8.6; group 3: 8.7-8.9; and group 4: 9.0-9.5). Group 1 had a lower LB rate (34.6% vs. 59.8-72.3%) and a higher rate of pregnancy loss (26% vs. 4.7-8.9%) compared with the other groups (p < 0.05). The receiver operating characteristic curve analysis verified that the iDAScore had a significant but limited ability to predict LB (area under the curve [AUC] = 0.64); this ability was significantly weaker than that of the combination of iDAScore, type of chromosomal abnormalities, and PI level (AUC = 0.67). In the comparison of the LB groups with the non-LB groups, the AI scores were significantly lower in the non-LB groups, both for euploid (median: 8.6 vs. 8.8) and mosaic (median: 8.0 vs. 8.6) SETs.

CONCLUSIONS

Although its predictive ability can be further enhanced, the AI score was significantly associated with LB probability in SET cycles. Euploid or mosaic blastocysts with low AI scores (≤ 7.8) were associated with a lower LB rate, indicating the potential of this annotation-free AI system as a decision-support tool for deselecting embryos with poor pregnancy outcomes following PGT-A.

Correlations between a deep learning-based algorithm for embryo evaluation with cleavage-stage cell numbers and fragmentation

RESEARCH QUESTION

Do cell numbers and degree of fragmentation in cleavage-stage embryos, assessed manually, correlate with evaluations made by deep learning algorithm model iDAScore v2.0?

DESIGN

Retrospective observational study (n = 5040 embryos; 1786 treatments) conducted at two Swedish assisted reproductive technology centres between 2016 and 2021. Fresh single embryo transfer was carried out on days 2 or 3 after fertilization. Embryo evaluation using iDAScore v2.0 was compared with manual assessment of numbers of cells and grade of fragmentation, analysed by video sequences.

RESULTS

Data from embryos transferred on days 2 and 3 showed that having three or fewer cells compared with four or fewer cells on day 2, and six or fewer cells versus seven to eight cells on day 3, correlated significantly with a difference in iDAScore (medians 2.4 versus 4.0 and 2.6 versus 4.6 respectively; both P < 0.001). The iDAScore for 0-10% fragmentation was significantly higher compared with the groups with higher fragmentation (P < 0.001). When combining cell numbers and fragmentation, iDAScore values decreased as fragmentation increased, regardless of cell number. iDAScore discriminated between embryos that resulted in live birth or no live birth (AUC of 0.627 and 0.607), compared with the morphological model (AUC of 0.618 and 0.585) for day 2 and day 3, respectively.

CONCLUSIONS

The iDAScore v2.0 values correlated significantly with cell numbers and fragmentation scored manually for cleavage-stage embryos on days 2 and 3. iDAScore had some predictive value for live birth, conditional that embryo selection was based on morphology.

Deep learning for embryo evaluation using time-lapse: a systematic review of diagnostic test accuracy

OBJECTIVE

This study aimed to investigate the accuracy of convolutional neural network models in the assessment of embryos using time-lapse monitoring.

DATA SOURCES

A systematic search was conducted in PubMed and Web of Science databases from January 2016 to December 2022. The search strategy was carried out by using key words and MeSH (Medical Subject Headings) terms.

STUDY ELIGIBILITY CRITERIA

Studies were included if they reported the accuracy of convolutional neural network models for embryo evaluation using time-lapse monitoring. The review was registered with PROSPERO (International Prospective Register of Systematic Reviews; identification number CRD42021275916).

METHODS

Two reviewer authors independently screened results using the Covidence systematic review software. The full-text articles were reviewed when studies met the inclusion criteria or in any uncertainty. Nonconsensus was resolved by a third reviewer. Risk of bias and applicability were evaluated using the QUADAS-2 tool and the modified Joanna Briggs Institute or JBI checklist.

RESULTS

Following a systematic search of the literature, 22 studies were identified as eligible for inclusion. All studies were retrospective. A total of 522,516 images of 222,998 embryos were analyzed. Three main outcomes were evaluated: successful in vitro fertilization, blastocyst stage classification, and blastocyst quality. Most studies reported >80% accuracy, and embryologists were outperformed in some. Ten studies had a high risk of bias, mostly because of patient bias.

CONCLUSION

The application of artificial intelligence in time-lapse monitoring has the potential to provide more efficient, accurate, and objective embryo evaluation. Models that examined blastocyst stage classification showed the best predictions. Models that predicted live birth had a low risk of bias, used the largest databases, and had external validation, which heightens their relevance to clinical application. Our systematic review is limited by the high heterogeneity among the included studies. Researchers should share databases and standardize reporting.

Association between a morphokinetic ploidy prediction model risk score and miscarriage and live birth: a multicentre cohort study

OBJECTIVE

To determine whether the aneuploidy risk score from a morphokinetic ploidy prediction model, Predicting Euploidy for Embryos in Reproductive Medicine (PREFER), is associated with miscarriage and live birth outcomes.

DESIGN

Multicentre cohort study.

SETTING

Nine in vitro fertilization clinics in the United Kingdom.

PATIENTS

Data were obtained from the treatment of patients from 2016-2019. A total of 3587 fresh single embryo transfers were included; preimplantation genetic testing for aneuploidy) cycles were excluded.

INTERVENTION

PREFER is a model developed using 8,147 biopsied blastocyst specimens to predict ploidy status using morphokinetic and clinical biodata. A second model using only morphokinetic (MK) predictors was developed, P PREFER-MK. The models will categorize embryos into the following three risk score categories for aneuploidy: "high risk," "medium risk," and "low risk."

MAIN OUTCOME MEASURES

The primary outcomes are miscarriage and live birth. Secondary outcomes include biochemical clinical pregnancy per single embryo transfer.

RESULTS

When applying PREFER, the miscarriage rates were 12%, 14%, and 22% in the "low risk," "moderate risk," and "high risk" categories, respectively. Those embryos deemed "high risk" had a significantly higher egg provider age compared with "low risk," and there was little variation in risk categories in patients of the same age. The trend in miscarriage rate was not seen when using PREFER-MK; however, there was an association with live birth, increasing from 38%-49% and 50% in the "high risk," "moderate risk," and "low risk" groups, respectively. An adjusted logistic regression analysis demonstrated that PREFER-MK was not associated with miscarriage when comparing "high risk" to "moderate risk" embryos (odds ratio [OR], 0.87; 95% confidence interval [CI], 0.63-1.63) or "high risk" to "low risk" embryos (OR, 1.07; 95% CI, 0.79-1.46). An embryo deemed "low risk" by PREFER-MK was significantly more likely to result in a live birth than those embryos graded "high risk" (OR, 1.95; 95% CI, 1.65-2.25).

CONCLUSION

The PREFER model's risk scores were significantly associated with live births and miscarriages. Importantly, this study also found that this model applied too much weight to clinical factors, such that it could no longer rank a patient's embryos effectively. Therefore, a model including only MKs would be preferred; this was similarly associated with live birth but not miscarriage.

Application of artificial intelligence in gametes and embryos selection

Gamete and embryo quality are critical to the success rate of Assisted Reproductive Technology (ART) cycles, but there remains a lack of methods to accurately measure the quality of sperm, oocytes and embryos. The ability of Artificial Intelligence (AI) technology to analyze large amounts of data, especially video and images, is particularly useful in gamete and embryo assessment and selection. The well-trained model has fast calculation speed and high accuracy, which can help embryologists to perform more objective gamete and embryo selection. Various artificial intelligence models have been developed for gamete and embryo assessment, some of which exhibit good performance. In this review, we summarize the latest applications of AI technology in semen analysis, as well as selection for sperm, oocyte and embryo, and discuss the existing problems and development directions of artificial intelligence in this field.

Making and selecting the best embryo in the laboratory

Over the past 4 decades our ability to maintain a viable human embryo in vitro has improved dramatically, leading to higher implantation rates. This has led to a notable shift to single blastocyst transfer and the ensuing elimination of high order multiple gestations. Future improvements to embryo culture systems will not only come from new improved innovative media formulations (such as the inclusion of antioxidants), but plausibly by moving away from static culture to more dynamic perfusion-based systems now made a reality owing to the breakthroughs in three-dimensional printing technology and micro fabrication. Such an approach has already made it feasible to create high resolution devices for intracytoplasmic sperm injection, culture, and cryopreservation, paving the way not only for improvements in outcomes but also automation of assisted reproductive technology. Although improvements in culture systems can lead to further increases in pregnancy outcomes, the ability to quantitate biomarkers of embryo health and viability will reduce time to pregnancy and decrease pregnancy loss. Currently artificial intelligence is being used to assess embryo development through image analysis, but we predict its power will be realized through the creation of selection algorithms based on the integration of information related to metabolic functions, cell-free DNA, and morphokinetics, thereby using vast amounts of different data types obtained for each embryo to predict outcomes. All of this will not only make assisted reproductive technology more effective, but it will also make it more cost effective, thereby increasing patient access to infertility treatment worldwide.

Embryo selection through artificial intelligence versus embryologists: a systematic review

STUDY QUESTION

What is the present performance of artificial intelligence (AI) decision support during embryo selection compared to the standard embryo selection by embryologists?

SUMMARY ANSWER

AI consistently outperformed the clinical teams in all the studies focused on embryo morphology and clinical outcome prediction during embryo selection assessment.

WHAT IS KNOWN ALREADY

The ART success rate is ∼30%, with a worrying trend of increasing female age correlating with considerably worse results. As such, there have been ongoing efforts to address this low success rate through the development of new technologies. With the advent of AI, there is potential for machine learning to be applied in such a manner that areas limited by human subjectivity, such as embryo selection, can be enhanced through increased objectivity. Given the potential of AI to improve IVF success rates, it remains crucial to review the performance between AI and embryologists during embryo selection.

STUDY DESIGN SIZE DURATION

The search was done across PubMed, EMBASE, Ovid Medline, and IEEE Xplore from 1 June 2005 up to and including 7 January 2022. Included articles were also restricted to those written in English. Search terms utilized across all databases for the study were: ('Artificial intelligence' OR 'Machine Learning' OR 'Deep learning' OR 'Neural network') AND ('IVF' OR 'in vitro fertili*' OR 'assisted reproductive techn*' OR 'embryo'), where the character '*' refers the search engine to include any auto completion of the search term.

PARTICIPANTS/MATERIALS SETTING METHODS

A literature search was conducted for literature relating to AI applications to IVF. Primary outcomes of interest were accuracy, sensitivity, and specificity of the embryo morphology grade assessments and the likelihood of clinical outcomes, such as clinical pregnancy after IVF treatments. Risk of bias was assessed using the Modified Down and Black Checklist.

MAIN RESULTS AND THE ROLE OF CHANCE

Twenty articles were included in this review. There was no specific embryo assessment day across the studies-Day 1 until Day 5/6 of embryo development was investigated. The types of input for training AI algorithms were images and time-lapse (10/20), clinical information (6/20), and both images and clinical information (4/20). Each AI model demonstrated promise when compared to an embryologist's visual assessment. On average, the models predicted the likelihood of successful clinical pregnancy with greater accuracy than clinical embryologists, signifying greater reliability when compared to human prediction. The AI models performed at a median accuracy of 75.5% (range 59-94%) on predicting embryo morphology grade. The correct prediction (Ground Truth) was defined through the use of embryo images according to post embryologists' assessment following local respective guidelines. Using blind test datasets, the embryologists' accuracy prediction was 65.4% (range 47-75%) with the same ground truth provided by the original local respective assessment. Similarly, AI models had a median accuracy of 77.8% (range 68-90%) in predicting clinical pregnancy through the use of patient clinical treatment information compared to 64% (range 58-76%) when performed by embryologists. When both images/time-lapse and clinical information inputs were combined, the median accuracy by the AI models was higher at 81.5% (range 67-98%), while clinical embryologists had a median accuracy of 51% (range 43-59%).

LIMITATIONS REASONS FOR CAUTION

The findings of this review are based on studies that have not been prospectively evaluated in a clinical setting. Additionally, a fair comparison of all the studies were deemed unfeasible owing to the heterogeneity of the studies, development of the AI models, database employed and the study design and quality.

WIDER IMPLICATIONS OF THE FINDINGS

AI provides considerable promise to the IVF field and embryo selection. However, there needs to be a shift in developers' perception of the clinical outcome from successful implantation towards ongoing pregnancy or live birth. Additionally, existing models focus on locally generated databases and many lack external validation.

STUDY FUNDING/COMPETING INTERESTS

This study was funded by Monash Data Future Institute. All authors have no conflicts of interest to declare.

REGISTRATION NUMBER

CRD42021256333.

In Contemporary Reproductive Medicine Human Beings are Not Yet Dispensable

In the past few years almost every aspect of an IVF cycle has been investigated, including research on sperm, color doppler in follicular studies, prediction of embryo cleavage, prediction of blastocyst formation, scoring blastocyst quality, prediction of euploid blastocysts and live birth from blastocysts, improving the embryo selection process, and for developing deep machine learning (ML) algorithms for optimal IVF stimulation protocols. Also, artificial intelligence (AI)-based methods have been implemented for some clinical aspects of IVF, such as assessing patient reproductive potential and individualizing gonadotropin stimulation protocols. As AI has the inherent capacity to analyze "Big" data, the goal will be to apply AI tools to the analysis of all embryological, clinical, and genetic data to provide patient-tailored individualized treatments. Human skillsets including hand eye coordination to perform an embryo transfer is probably the only step of IVF that is outside the realm of AI & ML today. Embryo transfer success is presently human skill dependent and deep machine learning may one day intrude into this sacred space with the advent of programed humanoid robots. Embryo transfer is arguably the rate limiting step in the sequential events that complete an IVF cycle. Many variables play a role in the success of embryo transfer, including catheter type, atraumatic technique, and the use of sonography guidance before and during the procedure of embryo transfer. In contemporary Reproductive Medicine human beings are not yet dispensable.

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.

Artificial Intelligence-Based Detection of Human Embryo Components for Assisted Reproduction by In Vitro Fertilization

Assisted reproductive technology is helping humans by addressing infertility using different medical procedures that help in a successful pregnancy. In vitro fertilization (IVF) is one of those assisted reproduction methods in which the sperm and eggs are combined outside the human body in a specialized environment and kept for growth. Assisted reproductive technology is helping humans by addressing infertility using different medical procedures that help in a successful pregnancy. The morphology of the embryological components is highly related to the success of the assisted reproduction procedure. In approximately 3-5 days, the embryo transforms into the blastocyst. To prevent the multiple-birth risk and to increase the chance of pregnancy the embryologist manually analyzes the blastocyst components and selects valuable embryos to transfer to the women's uterus. The manual microscopic analysis of blastocyst components, such as trophectoderm, zona pellucida, blastocoel, and inner cell mass, is time-consuming and requires keen expertise to select a viable embryo. Artificial intelligence is easing medical procedures by the successful implementation of deep learning algorithms that mimic the medical doctor's knowledge to provide a better diagnostic procedure that helps in reducing the diagnostic burden. The deep learning-based automatic detection of these blastocyst components can help to analyze the morphological properties to select viable embryos. This research presents a deep learning-based embryo component segmentation network (ECS-Net) that accurately detects trophectoderm, zona pellucida, blastocoel, and inner cell mass for embryological analysis. The proposed method (ECS-Net) is based on a shallow deep segmentation network that uses two separate streams produced by a base convolutional block and a depth-wise separable convolutional block. Both streams are densely concatenated in combination with two dense skip paths to produce powerful features before and after upsampling. The proposed ECS-Net is evaluated on a publicly available microscopic blastocyst image dataset, the experimental segmentation results confirm the efficacy of the proposed method. The proposed ECS-Net is providing a mean Jaccard Index (Mean JI) of 85.93% for embryological analysis.

Correlation between an annotation-free embryo scoring system based on deep learning and live birth/neonatal outcomes after single vitrified-warmed blastocyst transfer: a single-centre, large-cohort retrospective study

Propose

Does an annotation-free embryo scoring system based on deep learning and time-lapse sequence images correlate with live birth (LB) and neonatal outcomes?

Methods

Patients who underwent SVBT cycles (3010 cycles, mean age: 39.3 ± 4.0). Scores were calculated using the iDAScore software module in the Vitrolife Technology Hub (Vitrolife, Gothenburg, Sweden). The correlation between iDAScore, LB rates, and total miscarriage (TM), including 1st- and 2nd-trimester miscarriage, was analysed using a trend test and multivariable logistic regression analysis. Furthermore, the correlation between the iDAScore and neonatal outcomes was analysed.

Results

LB rates decreased as iDAScore decreased (P < 0.05), and a similar inverse trend was observed for the TM rates. Additionally, multivariate logistic regression analysis showed that iDAScore significantly correlated with increased LB (adjusted odds ratio: 1.811, 95% CI: 1.666–1.976, P < 0.05) and decreased TM (adjusted odds ratio: 0.799, 95% CI: 0.706–0.905, P < 0.05). There was no significant correlation between iDAScore and neonatal outcomes, including congenital malformations, sex, gestational age, and birth weight. Multivariate logistic regression analysis, which included maternal and paternal age, maternal body mass index, parity, smoking, and presence or absence of caesarean section as confounding factors, revealed no significant difference in any neonatal characteristics.

Conclusion

Automatic embryo scoring using iDAScore correlates with decreased miscarriage and increased LB and has no correlation with neonatal outcomes.

Supplementary information

The online version contains supplementary material available at 10.1007/s10815-022-02562-5.

Reporting on the Value of Artificial Intelligence in Predicting the Optimal Embryo for Transfer: A Systematic Review including Data Synthesis

Artificial intelligence (AI) has been gaining support in the field of in vitro fertilization (IVF). Despite the promising existing data, AI cannot yet claim gold-standard status, which serves as the rationale for this study. This systematic review and data synthesis aims to evaluate and report on the predictive capabilities of AI-based prediction models regarding IVF outcome. The study has been registered in PROSPERO (CRD42021242097). Following a systematic search of the literature in Pubmed/Medline, Embase, and Cochrane Central Library, 18 studies were identified as eligible for inclusion. Regarding live-birth, the Area Under the Curve (AUC) of the Summary Receiver Operating Characteristics (SROC) was 0.905, while the partial AUC (pAUC) was 0.755. The Observed: Expected ratio was 1.12 (95%CI: 0.26–2.37; 95%PI: 0.02–6.54). Regarding clinical pregnancy with fetal heartbeat, the AUC of the SROC was 0.722, while the pAUC was 0.774. The O:E ratio was 0.77 (95%CI: 0.54–1.05; 95%PI: 0.21–1.62). According to this data synthesis, the majority of the AI-based prediction models are successful in accurately predicting the IVF outcome regarding live birth, clinical pregnancy, clinical pregnancy with fetal heartbeat, and ploidy status. This review attempted to compare between AI and human prediction capabilities, and although studies do not allow for a meta-analysis, this systematic review indicates that the AI-based prediction models perform rather similarly to the embryologists’ evaluations. While AI models appear marginally more effective, they still have some way to go before they can claim to significantly surpass the clinical embryologists’ predictive competence.