Embryo Grading System Research Articles

O-248 Novel early cleavage parameters improve embryo selection and reach ploidy predictive values comparable to artificial intelligence

Abstract Study question Can we improve embryo selection and ploidy prediction using novel morphokinetic parameters in blastocyst development, and obtain results comparable to artificial intelligence? Summary answer Time and pattern of novel parameter st2 are related to ploidy. Including them in embryo evaluation improves ploidy prediction, reaching predictive values comparable to AI. What is known already Time-lapse has allowed us to detect more morphokinetic parameters and improve embryo grading systems. However, ploidy prediction still relies on invasive techniques to obtain a reliable result. Artificial intelligence is promising and can already predict blastulation or implantation with good results, but not ploidy yet. Moreover, AI presents some other limitations now, including the black box behind its decisions, which doesn’t allow clinicians to know what parameters are being considered the most. Improving the manual evaluation tools used by embryologists can help improve clinical outcomes until AI is fully introduced in IVF. Study design, size, duration A single center retrospective study was performed. Data included a total of 608 blastocyst from 106 treatments between January 2018 and July 2022. Participants/materials, setting, methods 160 treatments of patients for preimplantation genetic testing cycles were studied. Embryos were cultured in time-lapse incubators and were studied for both morphologic and kinetic parameters. 608 of the embryos developed to good quality blastocyst and were able to be biopsied, and therefore analyzed by PGT for aneuploidies on Day5 o Day6. Annotation of several morphokinetic characteristics was performed on all analyzed embryos. The study was blinded for ploidy and all clinical outcomes. Main results and the role of chance Improved embryo grading systems used by embryologist can reach high predictive values comparable to artificial intelligence. We present a new parameter, “st2, start of t2”, which refers to the first cytoplasmic movements detected at the beginning of the first cell cleavage, as highly implicated in ploidy status. Earlier st2 is associated to better euploidy rates (p < 0.001). Not only the time, but also the phenotype of the movements appears as related to ploidy. Embryos showing none or un-patterned random cytoplasmic movements show a marked tendency towards aneuploidy. Contrary, circular waves prior to cell division are highly associated with euploidy (90% of the cases) (p < 0.0001). The implication of the described parameters was also confirmed by logistic regression, with a receiver operating curve (ROC) value of 0.69 for ploidy prediction (95% confidence interval (CI), 0.62 to 0.76). We processed raw time-lapse images with an automatic annotation software using artificial intelligence and obtained a ploidy predictive value of 0.64. This means that using the optimal parameters in manual embryo evaluation, we are able to improve our grading systems and reach ploidy predictive values comparable than with AI. Limitations, reasons for caution This is a retrospective study. The ploidy predictive potential is now being tested on a prospective study. Wider implications of the findings Optimizing the indicators to select the most suitable blastocyst, like including st2, can help reducing the time until the pregnancy of an euploid baby, avoiding invasive and expensive methods. Moreover, by combining AI with the novel parameter st2 we could improve the potential of the ploidy predictions. Trial registration number not applicable

P-240 Is the intelligent Data Analysis Score (iDAScore) a reliable tool to predict pregnancy? Our center’s experience

Abstract Study question Is the intelligent Data Analysis Score (iDAScore) reliable in predicting a positive pregnancy test, clinical pregnancy and miscarriage rate after single vitrified-warmed blastocyst transfer (SVBT)? Summary answer There is a significant increase in positive pregnancy test and clinical pregnancy rates as iDAScore increases. The miscarriage rate is lower with increased iDAScore. What is known already With the improvement of assisted reproductive technology (ART), elective single embryo transfer has become a standard. Optimizing embryo selection is a major challenge to increase reproductive outcomes. The Gardner embryo grading system is widely adopted for the evaluation of blastocyst quality, but this tool is subject to inter- and intra-variability. In recent years, artificial intelligence has found diverse applications in healthcare. Based on time-laps sequences of > 180’000 developing embryos, iDAScore is a new deep-learning model for objective and standardized blastocyst scoring. However, its reliability in predicting pregnancy and miscarriage rates remains unknown. Study design, size, duration This monocentric retrospective pilot study was conducted from May 2023 to December 2023. A total of 206 homologous SVBT cycles from 206 patients were analyzed (one patient, one cycle). Exclusion criteria included: cycles involving preimplantation genetic testing, multiple and/or fresh blastocyst transfer, and blastocysts obtained after testicular sperm extraction. All vitrified-warmed blastocysts were scored using iDAScore version 1.0 model, ranging from 1 – 9.9. Participants/materials, setting, methods Scores were stratified into 3 groups: high-score (9.9 – 7.7), medium-score (7.6 – 5.5) and low-score (5.4 – 3.1). For each group, positive pregnancy test, clinical pregnancy and miscarriage rates were analyzed. SVBT cycles were then stratified into four maternal age groups (< 35, 35-37, 38-40, >41 years) and the same outcomes were analyzed. χ2 was used for statistics analysis. Main results and the role of chance Mean maternal age was 36.8 ± 4.9. Among the 206 vitrified-warmed blastocysts, 122 were scored as high-score, 57 as medium-score and 27 as low-score. Positive pregnancy test rate was significantly higher with increased iDAScore with 58% in the high-score group, 36.8% in the medium-score group and 17% in the low-score group (p < 0.0001). Clinical pregnancy rate, defined as the visualization of a fetal heartbeat at ultrasound scan, was also significantly higher with increased iDAScore, with 51.6% in the high-score group, 31.6% in the medium-score group and 17% in the low-score group (p < 0.0001). The miscarriage rate was lower with increased iDAScore, with 8.1% in the high-score group, 16.6% in the medium-score group and 50% in the low-score group, but this difference was not statistically significative (p 0.032). When adjusted to age group however, the miscarriage rate was found to be significantly lower in the high-score group when compared to the medium-score group (p < 0.0001); statistical analysis was not achieved in the low-score group due to the small sample size. Limitations, reasons for caution The major limitation of our study is its small sample size. Data collection is ongoing, to increase the study power. Additionally, many other factors, such as genetic, hormonal, metabolic and/or immunological, may play a role in achieving pregnancy. Wider implications of the findings Our results indicate that iDAScore may be a strong predictor of pregnancy and could be used as a complementary tool to the Gardner embryo grading system in the decision-making process of which blastocyst to transfer. Trial registration number N/A

P-505 Clinical rescue rate of abnormally fertilized oocytes and the relationship with patient, embryo and laboratory features

Abstract Study question Which are the main patient, embryonic and laboratory variables associated with a diploid result in abnormally fertilized embryos? Summary answer Embryo diploid constitution is associated with maternal age, trophectoderm and inner cell mass quality as well as pronuclear (PN) number predicted by time-lapse incubator. What is known already In conventional in vitro fertilization cycles, morphological PN evaluation can identify proper fertilization and predict embryo ploidy constitution. Atypical PN patterns like 0PN, 1PN or more than 2PN are associated with ploidy abnormalities. However, we have previously shown that PN number is insufficient to predict embryo ploidy composition, and advanced single nucleotide polymorphism (SNP)-based methodologies for preimplantation genetic testing (PGT) detect a significant number of genetically diploid embryos from presumed abnormally fertilized oocytes. In this study, we characterized the diploid rescue rate associated with abnormal PN configurations, uncovering the link between ploidy results and relevant patient, embryonic and clinical/laboratory parameters. Study design, size, duration Retrospective observational study evaluating the association between diploid rescue rate of abnormally fertilized embryos and different patient, embryonic and laboratory features. Embryos were cultured in conventional or EmbryoScope™ time-lapse incubator (Vitrolife) until blastocyst stage and classified according to Gardner embryo grading system. Ploidy assessment was performed on 343 euploid trophectoderm (TE) biopsies collected between January 2021 and November 2023. Results were stratified according to patients age, embryo quality parameters and different laboratory procedures. Participants/materials, setting, methods Multi-center international study involving consenting patients using autologous gametes and undergoing PGT for aneuploidies (PGT-A) combined with ploidy analysis. Ploidy protocol was performed using a validated targeted NGS approach composed of a custom panel of 357 SNPs and a proprietary pipeline of analysis. Reliable ploidy classification was obtained using a combined algorithm strategy based on SNPs allele frequencies. Categorical variables were expressed in percentages. Nominal variables were analysed using the Cochran-Armitage test to detect trends. Main results and the role of chance Among 343 euploid embryos, diploidy rates obtained for 0PN, 1PN, 2.1PN and 3PN groups were 86.8% (n = 118/136;95%CI=79.89-91.96) 43.7% (n = 59/135;95%CI=35.19-52.50), 65.0% (n = 13/20;95%CI=40.78-84.61) and 26.9% (n = 14/52;95%CI=15.57-41.02) respectively. Regarding patient parameters, advanced maternal age, but not paternal age, was significantly associated with lower diploid rescue rates, with a decreasing trend from 72.5% (n = 79/109;95%CI=63.10-80.60) below age 35 to 42.9% (n = 30/70;95%CI=31.09-55.25) above age 40 (p < 0.01). When considering embryo morphological evaluations, Cochran Armitage test revealed a significant link between diploidy rates and TE quality with a decreasing trend going from 72.1% (n = 80/111;95%CI=62.76-80.17) to 52.7% (n = 79/150;95%CI=44.36-60.87) and 45.8% (n = 22/48;95%CI=31.37-60.83) for embryos classified with TE grade as A, B, and C, respectively (p-value<0.01). Similarly, a decreasing trend was reported in relation to inner cell mass (ICM) quality, when moving from grade A (67,1%, n = 94/140;95%CI=58.23-75.33) to grade C (54.8%, n = 17/31;95%CI=36.03-72.68) (p = 0.03). No clear association was reported for embryo expansion degree. Finally, among laboratory variables (oocyte freezing procedure, fertilization method, incubator type), only time-lapse significantly impacted expected diploid results. Overall, if abnormal PN number is confirmed by time-lapse (n = 42 samples), the accuracy in predicting ploidy abnormalities increased from 29.2% (n = 82/281;95%CI=23.93-34.87), in conventional incubators, to 83.3% (n = 35/42;95%CI=68.64-93.03) (p < 0.01). Consequently, expected diploid rescue rate decreased from 65.5% (n = 184/281;95%CI=59.60-71.03) to 16.7% (n = 7/42;95%CI=6.97-31.36) (p < 0.01). Limitations, reasons for caution Parental DNA analysis should be included to rule-out the residual risk of uniparental disomy in diploid embryos. The multi-center nature of the study didn’t support a uniform PN annotation, especially for centers lacking time-lapse observation. Future studies should focus on the clinical follow-up after the transfer of rescued euploid-diploid embryos. Wider implications of the findings We believe we reported the first evidence of maternal age effect on ploidy results in the largest dataset of abnormally fertilized embryos. We revealed a link between ploidy constitution and blastocyst morphological quality, providing important information for patient counselling. Time-lapse incubators offer a more accurate identification of abnormally fertilized oocytes. Trial registration number n/a

Accurate system for assessment and selection of human embryo after in vitro fertilization

In Vitro Fertilization (IVF) is a complex medical procedure designed to assist individuals in overcoming infertility by facilitating the union of sperm and egg outside the human body. Not all embryos created during the IVF process have the same potential for successful implantation and development. Embryo grading allows embryologists to assess the quality of embryos based on specific criteria, such as cell division, symmetry, and the presence of fragmentation. By assigning grades, they can identify the most viable embryos with the highest likelihood of successful implantation. The conventional methods for embryo grading are time-consuming, labor-intensive, and prone to errors due to their subjective nature. Therefore, there is a pressing need for the development of an accurate and automated grading system to address these challenges. In this study, I propose a human embryo grading system based on machine learning. The proposed system utilizes embryo images as inputs and produces probabilities to predict the success or failure of each embryo. I utilize a convolutional neural network to develop the grading model and introduce a novel hierarchical feature extraction approach to enhance accuracy. Through extensive experiments, I demonstrate that the proposed method surpasses previous approaches with a significant performance margin.

A simplified grading system for in vivo and in vitro derived vitrified equine embryos

The ability to predict the viability of embryos before vitrification and thawing has important commercial applications in any breeding program. The aim of this study was to develop and evaluate a simplified embryo grading system for both in vivo- and in vitro-derived vitrified day 8 embryos. The in vivo derived (n=109) and in vitro – intracytoplasmic sperm injection derived (n=145) embryos were produced in a commercial embryo program. The embryos were classified as Grade 1, 2 or 3 based on the amount of extruded material between the trophoblast and the zona pellucida observed during the vitrification process. The embryos were vitrified at day 8 of development in a two-step system with increasing concentrations of dimethylsulfoxide and ethylene glycol and 0.5 M sucrose in the final solution. Each embryo was thawed in 0.3 M and then 0.15 M sucrose before transfer into holding medium for non-surgical transfer into a recipient mare. Analysis of the relationship between the embryo grading system and pregnancy rates after vitrification, thawing and transfer of in vivo and in vitro derived embryos confirmed that there was a significant effect of origin (in vivo vs in vitro; P ≤ 0.05), and embryo grade (P ≤ 0.001) on embryo survival after transfer. In conclusion, this simplified grading system is predictive of embryo survival for both in vivo- and in vitro- derived vitrified equine embryos.

The relationship between grade and metabolism for in vivo and ICSI derived embryos

The objective of this study was to describe the metabolism of glucose, pyruvate and lactate by in vitro- (ICSI derived) and in vivo-derived equine embryos and the relationship with embryo grade and sex. During the course of this study, a simplified embryo grading system was developed for both in vivo- and in vitro-derived embryos that proved to be significantly associated with both embryo metabolism and survival rates following transfer into a recipient mare. A total of 15 in vivo derived embryos and 32 ICSI-derived embryos were cultured in protein free synthetic oviduct fluid (SOF) for 24h between day 8 and day 9 of development. Aliquots of 10µL of media before culture (t=0) and spent media after culture (t=24h) were snap frozen for subsequent analysis by GC-Mass Spectrometry for glucose, lactate and pyruvate. The embryos were photographed and classified as grade 1, 2 or 3 based on the percentage of extruded debris between the trophoblast and the zona pellucida. In this study the sex of the embryos (in vivo n = 5, ICSI n = 25) was determined by PCR. At day 8 of development, the embryo metabolism of glucose, pyruvate and lactate in relation to embryo origin (in vivo vs in vitro), grade and sex wasevaluated following 24h of culture in protein free media. Overall, there was a positive correlation between embryo volume and lactate peak area (r2 = 88.2%, p ≤ 0.001) and a negative correlation with both glucose peak area (r2 = 29.8%, p =0.023) and pyruvate peak area (r2 = 46.2%, p =0.003) in the spent media. These results illustrated that as embryo volume increased, the amount of glucose and pyruvate in the media decreased and the production of lactate increased. After accounting for embryo volume, general linear model analysis revealed an effect of embryo grade (p =0.031), but no significant effect of embryo sex or origin on lactate consumption over 24h. Grade 2 embryos, however, consumed more lactate than Grade 1 embryos (p = 0.029). In conclusion, the metabolic relationships between glucose, lactate and pyruvate were described for both in vivo- and in vitro-derived equine embryos in this study. Understanding the optimal metabolic conditions for in vitro-derived embryos in comparison with in vivo-derived embryos will enable the development of models to investigate early embryonic death and to optimize pregnancy rates in the mare.

Towards Automation in IVF: Pre-Clinical Validation of a Deep Learning-Based Embryo Grading System during PGT-A Cycles.

Preimplantation genetic testing for aneuploidies (PGT-A) is arguably the most effective embryo selection strategy. Nevertheless, it requires greater workload, costs, and expertise. Therefore, a quest towards user-friendly, non-invasive strategies is ongoing. Although insufficient to replace PGT-A, embryo morphological evaluation is significantly associated with embryonic competence, but scarcely reproducible. Recently, artificial intelligence-powered analyses have been proposed to objectify and automate image evaluations. iDAScore v1.0 is a deep-learning model based on a 3D convolutional neural network trained on time-lapse videos from implanted and non-implanted blastocysts. It is a decision support system for ranking blastocysts without manual input. This retrospective, pre-clinical, external validation included 3604 blastocysts and 808 euploid transfers from 1232 cycles. All blastocysts were retrospectively assessed through the iDAScore v1.0; therefore, it did not influence embryologists' decision-making process. iDAScore v1.0 was significantly associated with embryo morphology and competence, although AUCs for euploidy and live-birth prediction were 0.60 and 0.66, respectively, which is rather comparable to embryologists' performance. Nevertheless, iDAScore v1.0 is objective and reproducible, while embryologists' evaluations are not. In a retrospective simulation, iDAScore v1.0 would have ranked euploid blastocysts as top quality in 63% of cases with one or more euploid and aneuploid blastocysts, and it would have questioned embryologists' ranking in 48% of cases with two or more euploid blastocysts and one or more live birth. Therefore, iDAScore v1.0 may objectify embryologists' evaluations, but randomized controlled trials are required to assess its clinical value.

P-284 Correlation between a new day 4 embryo scoring system and blastocyst development and pregnancy rate

Abstract Study question Does a new day 4 embryo scoring system predict blastocyst development and aneuploidies? Summary answer The new day 4 embryo scoring system correlates with blastocyst development, blastocyst morphology, day of blastulation, and pregnancy rate. What is known already A small number of studies propose different day 4 embryo grading systems. The majority of them report no differences in implantation and clinical pregnancy outcome between day 4 and blastocyst transfer. Moreover, only limited data exist on the association between day 4 grading systems, embryo development, and PGT-A results. In addition, none of the previous studies analyzed the totality of the different day 4 embryo morphology and their subsequent development. Study design, size, duration This is a single-center retrospective analysis from June 2019 to August 2021 including 762 embryos from 129 cycles. Day 4 embryos were graded A to J considering A the best morphology and were classified into 1, 2, 3, or 4 categories (1 being the more advanced development). Categories were compared among them to assess blastocyst development and aneuploidy rate. Participants/materials, setting, methods Data were obtained from medical records with a previously given consent. Patient characteristics such as age, body mass index (BMI), and anti-Mullerian hormone (AMH) were compared among the four categories using ANOVA with Multiple Comparisons. Blastocyst development, blastocyst morphology, day of blastulation, and pregnancy rate were compared among the categories using Chi-square contingency. All analyses considered a significant p-value of 0.05. Main results and the role of chance Younger women presented significantly more category 1 of day 4 embryos when compared with the other categories (p < 0.0001). BMI and AMH were similar among the categories (p = 0.3063 and p = 0.2533, respectively). There was a high correlation between day 4 categories and PGT-A blastocyst development (p < 0.0001), blastocyst morphology (p < 0.0001), and day of blastulation (p < 0.0001). Also, there was a significant correlation between pregnancy rates and the four categories (p = 0.0192). No correlation was observed when aneuploidy rate and level of aneuploidy were analyzed (p = 0.3614 and p = 0.1490, respectively). Limitations, reasons for caution The limited number of embryos analyzed is a limitation of the study, and additional data are being gathered. Multiple embryos of the same patients can be a cofounder of the study and more robust statistical analysis may overcome it. Wider implications of the findings The new day 4 embryo score may be useful for embryo selection to PGT-A biopsy. Moreover, the proposed score may help to add valuable information to select the best embryos for transfer. Trial registration number not applicable

O-122 Reporting on the value of Artificial Intelligence in predicting the optimal embryo for transfer: A systematic review and meta-analysis

Abstract Study question Are Artificial Intelligence (AI) based models effective in robustly predicting in vitro fertilization (IVF) outcome by assessing embryo quality? Summary answer The majority of the AI-based models could provide an accurate prediction regarding live birth, clinical pregnancy, clinical pregnancy with fetal heartbeat and embryo ploidy status. What is known already Precision and consistency in embryo quality evaluation are of paramount importance regarding the outcome of an IVF cycle. Numerous embryo grading and evaluation systems, employing morphological and morphokinetical assessment, have been proposed but without reaching a consensus yet. The main limitation of the aforementioned assessment systems is that they depend on human evaluation, which may be subject to subjectivity and interobserver variation. Thus, automated prediction models may be essential to optimize objectivity and reliability of embryo grading. Artificial neural network models may process microscopy images or time-lapse videos as input to predict the embryos’ potential competency. Study design, size, duration A systematic review and meta-analysis including 18 published studies. The population consists of preimplantation embryos suitable for embryo transfer in IVF/ICSI cycles following employment of an AI-based prediction model. The outcome measures are prediction of live birth, clinical pregnancy, clinical pregnancy with heartbeat and ploidy status. Participants/materials, setting, methods A systematic search of the literature was performed in the databases of Pubmed/Medline, Embase, and Cochrane Central Library limited to articles published in English up to August 2021. The initial search yielded a total of 694 studies with 97 of them being duplicates and other 579 being excluded on the grounds of not fulfilling inclusion criteria. Following full-text screening and citation mining a total of 18 studies were identified to be eligible for inclusion. Main results and the role of chance Four studies reported on prediction of live birth. The sensitivity was 70.6% (95%C.I.: 38.1-90.4%) and specificity was 90.6% (95%C.I.:79.3-96.1%). The Area Under the Curve (AUC) of the Summary Receiver Operating Characteristics (SROC) curve was 0.905, while the partial AUC (pAUC) was 0.755. Employing the Bayesian approach, the total Observed:Expected ratio (O:E) was 1.12 (95%CI: 0.26–2.37; 95%PI:0.02-6.54). Ten studies reported on prediction of clinical pregnancy. The sensitivity and the specificity were 71% (95%C.I.: 58.1-81.2%) and 62.5% (95%C.I.: 47.4-75.5%) respectively. The AUC was 0.716, while pAUC was 0.693. Moreover, the total O:E ratio was 0.92 (95%CI: 0.61–1.28; 95%PI:0.13-2.43). Eight studies reported on prediction of clinical pregnancy with fetal heartbeat the sensitivity was 75.2% (95%C.I.: 66.8-82%) and the specificity was 55.3% (95%C.I.: 41.2-68.7%). The AUC 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). Four studies reported on the ploidy status of the embryo. The sensitivity and specificity were 59.4% (95%C.I.: 45.0-73.1%) and 79.2% (95%C.I.: 70.1-86.1%) respectively. The AUC was 0.751 and the pAUC was 0.585. The total O:E ratio was 0.86 (95%CI: 0.42 – 1.27; 95%PI: 0.03-1.83). Limitations, reasons for caution The limited number of studies fulfilling inclusion criteria, along with the different designs applied when developing AI models which may lead to increased heterogeneity, stand as limitations. Inclusion of women regardless of their age presents as another limitation, as advanced maternal age has been associated with diminished IVF outcomes. Wider implications of the findings Albeit, our findings support that AI is a highly promising tool in the era of personalized medicine providing precise predictions it does not appear to considerably surpass human prediction capabilities. More studies and more collaborations between the developers are of paramount importance prior to AI becoming the gold standard. Trial registration number Not applicable

Non-invasive embryo selection strategy for clinical IVF to avoid wastage of potentially competent embryos

Research questionCan a non-invasive embryo transfer strategy provide a reference for embryo selection to be established? DesignChromosome sequencing of 345 paired blastocyst culture medium and whole blastocyst samples was carried out and a non-invasive embryo grading system was developed based on the random forest machine learning algorithm to predict blastocyst ploidy. The system was validated in 266 patients, and a blinded prospective observational study was conducted to investigate clinical outcomes between machine learning-guided and traditional non-invasive preimplantation genetic testing for aneuploidy (niPGT-A) analyses. Embryos were graded as A, B or C according to their euploidy probability levels predicted by non-invasive chromosomal screening (NICS). ResultsHigher live birth rate was observed in A- versus C-grade embryos (50.4% versus 27.1%, P = 0.006) and B- versus C-grade embryos (45.3% versus 27.1%, P = 0.022) and lower miscarriage rate in A- versus C-grade embryos (15.9% versus 33.3%, P = 0.026) and B- versus C-grade embryos (14.3% versus 33.3%, P = 0.021). The embryo utilization rate was significantly higher through the machine learning strategy than the conventional dichotomic judgment of euploidy or aneuploidy in the niPGT-A analysis (78.8% versus 57.9%, P < 0.001). Better outcomes were observed in A- and B-grade embryos versus C-grade embryos and higher embryo utilization rates through the machine learning strategy compared with traditional niPGT-A analysis. ConclusionA machine learning guided embryo grading system can be used to optimize embryo selection and avoid wastage of potential embryos.

P–267 Characterising a novel embryo grading system for 4-cell embryos including symmetry, fragmentation, cell configuration, cell contacts per cell, distance between cells, and cell adhesion strength

Abstract Study question What other cell junction factors can be easily characterised from imagery of 4-cell embryos to assist in embryo classification and prediction of viability? Summary answer 4-cell embryo grading should not only account for symmetry and fragmentation, but also cell configuration and cell adhesion quality. What is known already 4-cell embryos are clinically classified according to cell number, symmetry and fragmentation without accounting for cell orientation or quality of cell junctions. Our previous work has focused on classification of 4-cell embryos according to overall embryo shape (“tetrahedral” versus “planar”) using artificial intelligence. Our work, as confirmed by others, has demonstrated that embryo shape at the 4-cell stage is an important determinant of the ability of the embryo to reach blastocyst, utilisation, pregnancy and live birth. This is thought to be because of variations in intracellular communication between cells in embryos with different orientations, and consequently, different intracellular junction phenotypes. Study design, size, duration Using geometrical principles, possible permutations of 4-cell embryos (excluding redundant mirrored permutations) were identified and further classified based on shape and number of cell junctions. For ease of calculations, cells were assumed to be spherical, with at least one intracellular cell contact and symmetrical in size with other cells in the same embryo. Participants/materials, setting, methods The six distances between centroids of permutations for each configuration were calculated relative to the size of the cell, and the shortest distance between cell membranes. Adhesion was characterised from embryo imagery based on the overall shape of the cell, external angle between cells and the length of cell contact (the more spherical the cell, the larger the angle and the longer the cell contact point, the stronger the adhesion, adapted from Winklbauer,2015). Main results and the role of chance 4-cell embryos may be classified into 13 variant configurations: 1 typical Tetrahedral, 2 quasi-tetrahedral, 10 planar. These variants were classified according to number of cells with 0,1,2 and 3 intracellular contacts, leading to six possible configurations: 0004(tetrahedral), 0022(quasi tetrahedral/planar), 0040 (planar), 0121(quasi tetrahedral/planar), 0301(planar), 0220(planar). The number of total cell junctions in the embryo in each of these configurations was 12,10,8,8,6,6 respectively, with tetrahedral embryos (0004) having twice the cell contacts compared to planar embryos (p &amp;lt; 0.001). Tetrahedral embryos have an advantage over the other embryo configurations in terms of better embryo communication, as demonstrated by the lower average and variation in distance and shorter sum of all intracellular distances between centroids (mean: 0.78 vs 0.94,0.98,1.04,1.09,1.19; stdev:0.06 vs 0.2,0.3,0.3,0.4,0.4,0.3; sum:4.7 vs 5.2,6.3,6.1,6.7,7.0,6.0 cell lengths) and between cells (mean: 0 vs 0.34,0.07,0.39,0.42,0.56; stdev: 0 vs 0.56,0.11,0.51,0.53,0.66; sum: 0 vs 2.71,0.43,3.13,3.38,4.46 cell lengths) observed in tetrahedral embryos versus other five configurations respectively (p &amp;lt; 0.001). Cell junctions were classified according to degree of cell adhesion: A:none (cells remain spherical in shape); B:weak (external angle between the cells is acute, there is a narrow visible cell junction); C:strong (external angle between the cells is obtuse with a wide visible cell junction). Limitations, reasons for caution: Follow-up studies will evaluate the impact of different cell shapes,cells without intracellular contact,and asymmetrical embryos. The proposed classification will be validated against a database of known outcome from 8 clinics from 6 countries to quantify the clinical implications of this classification,and the consistency of assessment by humans and AI . Wider implications of the findings: It is clear that differences in intracellular communication between cells in embryos with different orientations, and different intracellular junction phenotypes is an important determinant of embryo viability. Our classification system allows for an easy to use and mathematically sound criteria for classifying 4-cell embryo cell junction quality. Trial registration number NA

A New Day 4 Grading System to Assess Embryo Quality in Frozen Embryo Transfer Cycles

To present a new day 4 (D4) embryo grading system to assess embryos in frozen-thawed embryo transfer (FET) cycles. A new grading system (grades A-E) was developed from the 2011 ESHRE Istanbul Consensus for D4 embryos in FET cycles. Embryos with complete compaction were classified as grade A; those with partial compaction were assigned as grade B; and those without compaction were classified as grades C, D, and E according to their different blastomere number ratio (BNR; number of embryo blastomeres on D4/number of embryo blastomeres on D3, D4/D3). Embryos with a BNR of ≥ 1.5 were defined as grade C, those with a BNR of ≥ 1.2 and < 1.5 were defined as grade D, and those with a BNR of ≥ 1.0 and < 1.2 were defined as grade E. Using this proposed grading model, 5460 embryos with known implantation data were retrospectively analyzed after D4 transfer in FET cycles. The transferred embryos exhibited a similar declining trend in implantation and live birth rates from the top grade A to the lowest grade E. The in vitro fertilization group showed increased implantation rates of grade B and E embryos compared with the intracytoplasmic sperm injection group (grade B: 41.99%, 34.63%, χ2 = 5.84, p < 0.05 and grade E: 18.98%, 14.08, χ2 = 75.62, p < 0.01). Receiver-operating characteristic analysis revealed that our proposed model predicted the implantation outcomes and live birth rates of all embryos (area under the curve = 0.65; 95% confidence interval [CI],0.63-0.66; p < 0.01 and AUC = 0.73; 95%CI, 0.65-0.84; p < 0.001, respectively). This study demonstrates that the new grading system provided by us can be a useful tool for assisting embryo selection via changes in embryo morphology. D4 embryo transfer provides a simple and applicable method for FET cycles in daily practice.

Associations between a new day 4 embryo grading system and implantation rates in frozen embryo transfer (FET) cycles.

The purpose of this study is to present a new day 4 (D4) embryo grading system for the assessment of embryos in frozen-thawed embryo transfer (FET) cycles.A new grading system (grades A–E) was modified from the 2011 ESHRE Istanbul Consensus for D4 embryos in FET cycles. In total, we retrospectively analyzed 5640 embryos with known implantation data after D4 transfer in FET cycles by using this proposed grading model.The transferred embryos exhibited a similar declining trend in implantation rates from the top grade A to the lowest grade E. The implantation rates of grade B and E embryos in the in vitro fertilization group were significantly higher than that in the intracytoplasmic sperm injection group (grade B: 41.82%, 35.23%, χ2 = 5.85, P < .05 and grade E: 18.53%, 14.81, χ2 = 76.86, P < .01, respectively). The receiver operating characteristic analysis showed that our proposed model predicted the implantation outcomes of all embryos (area under the ROC curve = 0.65; 95% CI, 0.63–0.66; P < .01).This study demonstrated that the new grading system provided by us turned out to be a useful tool in assisting embryo selection via embryo morphological changes, and D4 embryo transfer provided a simple and applicable method for a daily routine in FET cycles.

Time to blastulation is superior to individual components of embryo grading for live-birth prediction

ObjectiveTo compare components of the embryo grading system with time for blastocyst formation (i.e., day of embryo transfer) for predicting live-birth rate in frozen embryo transfer cycles.DesignRetrospective cohort study.SettingUniversity-affiliated fertility clinic.Patient(s)From January 2015 to October 2018, 870 frozen embryos transferred in a total of 509 women and 728 cycles at our institution.Intervention(s)None.Main Outcome Measure(s)Probability of live birth per cycle.Result(s)In unadjusted analysis of embryo grading components, both inner cell mass (ICM) and trophectoderm grades demonstrated a correlation with live-birth rates. However, this effect was lost in the ICM subgroup analysis by day of embryo transfer and preserved only in declining trophectoderm grades of day-6 transfers. In the adjusted analysis for prediction of live birth, only day of transfer was statistically significant. When assessing the composite score by Society for Assisted Reproductive Technology (SART) embryo grading, good embryos that blastulated on day 6 were statistically significantly less likely than day-5 embryos to result in live birth (risk ratio 0.70; 95% confidence interval, 0.58–0.85). Finally, in a predictive model adjusted for all individual components of embryo grade, the day of blastulation was the only statistically significant contributor.Conclusion(s)Time to blastulation is superior to other individual components of embryonic grading for prediction of live birth.

Using Deep Learning with Large Dataset of Microscope Images to Develop an Automated Embryo Grading System

The assessment of embryo viability for in vitro fertilization (IVF) is mainly based on subjective visual analysis, with the limitation of intra- and inter-observer variation and a time-consuming task. In this study, we used deep learning with large dataset of microscopic embryo images to develop an automated grading system for embryo assessment. This study included a total of 171,239 images from 16,201 embryos of 4,146 IVF cycles at Stork Fertility Center (https://www.e-stork.com.tw) from March 6, 2014 to April 13, 2018. The images were captured by inverted microscope (Zeiss Axio Observer Z1) at 112 to 116 hours (Day 5) or 136 to 140 hours (Day 6) after fertilization. Using a pre-trained network trained on the ImageNet dataset as convolution base, we applied Convolutional Neural Network (CNN) on embryo images, using ResNet50 architecture to fine-tune ImageNet parameters. The predicted grading results was compared with the grading results from trained embryologists to evaluate the model performance. The images were labeled by trained embryologists, based on Gardner’s grading system: blastocyst development ranking from 3–6, ICM quality as A, B, or C; and TE quality as a, b, or c. After pre-processing, the images were divided into training, validation, and test groups, in which 60% were allocated to the training group, 20% to the validation group, and 20% to the test group. The ResNet50 algorithm was trained on the 60% images allocated to the training group, and the algorithm’s performance was evaluated using the 20% images allocated to the test group. The results showed an average predictive accuracy of 75.36% for the all three grading categories: 96.24% for blastocyst development, 91.07% for ICM quality, and 84.42% for TE quality. To the best of our knowledge, this is the first study of an automatic embryo grading system using large dataset from Asian population. Combing the promising results obtained in this study with time-lapse microscope system integrated with IVF Electronic Medical Record platform, a fully automated and non-invasive pipeline for embryo assessment will be achieved.

Late embryo cleavage as an indicator of chromosome aneuploidy by pre-implantation genetic screening

During in vitro fertilization (IVF), not all patients receive good quality oocytes or embryos. This study investigated the frequency of aneuploidy and mosaicism in embryos obtained from parthenogenetic oocytes and poor morphological embryos that also exhibited late cleavage development by preimplantation genetic screening (PGS). Twenty late cleavage embryos under IVF treatment were cultured until day 5, and one of the blastomeres was biopsied. The embryos were divided into three groups (parthenogenetic oocytes and poor-quality embryos with dark cytoplasm, several vacuoles or fragments in the cytoplasm; arrested embryos; and good-quality embryos). All samples were tested for abnormalities using array-comparative genomic hybridization (a-CGH). Twenty late cleavage-stage embryos were biopsied and analyzed for aneuploidy. A total 17 of 20 (85%) embryos were aneuploid and three (15%) were euploid. The incidence of aneuploidy in good-quality embryos was 75%, while poor-quality embryos displayed a higher frequency (86.7%) of aneuploidy. The parthenogenetic oocytes were invariably aneuploid (100%). Based on the embryo-grading system by Veeck, embryo grades 1 and 2 exhibited a lower incidence of aneuploidy (75%), while grades 3 and 4 showed 1.33-times higher aneuploidy rates (RR = 1.33; p < 0.001). The numerical defects of the chromosomes increased with the percent of fragmentation, with fragmented embryos showing 1.32-times greater aneuploidy rates than the non-fragmented ones. Embryos with uneven blastomeres showed 1.07-times higher aneuploidy rate when compared with embryos showing even blastomeres (p < 0.0064). Late embryo cleavage results in a higher incidence of aneuploidy and mosaicism upon comparison with early embryo cleavage.

A Review of The Society for Assisted Reproductive Technology Embryo Grading System and Proposed Modification.

The Society for Assisted Reproductive Technology (SART) method of embryo grad- ing is unique, simple, and widely practiced, and its use has been mandatory for SART membership programs since 2010. Developed by SART in 2006, the current embryo grading system categories, "good, fair, and poor," are limited because they do not describe the best 1-2 embryos in the interest of keeping pace with the shift in clinical practice to be more selective and to transfer fewer embryos. This inspired us to conduct a review on the SART embryo grading system. In this retrospective study, the literature on evaluation of human embryo quality in gen- eral, and the SART method of evaluation in particular, were reviewed for the period of 2000 to 2014. A multifaceted search pertaining to methods of embryo grading and trans- fer using a combination of relevant terms [embryo, mammalian, embryo transfer, grade, grading, morphology, biomarkers, SART, and in vitro fertilization (IVF)] was performed. The inclusion and exclusion in this review were dictated by the aim and scope of the study. Two investigators independently assessed the studies and extracted information. A total of 61 articles were reviewed. Very few studies have evaluated the efficacy of the SART embryo grading method. The present study suggests the necessity for revision of the current SART grading system. The system, as it is now, lacks criteria for describing the cohort specific best embryo and thus is of limited use in single embryo transfer. The study foresees heightened descriptive efficiency of the SART system by implementing the proposed changes. Strengths and weaknesses of the SART embryo grading were identified. Ideas for selecting the best cohort-specific embryo have been discussed, which may trigger methodological improvement in SART and other embryo grading systems.

Evaluation of a high implantation potential (HIP) embryo grading system designed to reduce multiple pregnancy

AimsAlthough we have progressed a long way in performing in vitro fertilization, one hurdle that still remains is the assessment of embryos. The aim of the current paper was to evaluate an embryo selection protocol based on the presence of high implantation potential (HIP) embryos on Day 3 embryo transfer outcomes. MethodsThe analysis of cycles from 6573 IVF cycles in a private fertility clinic was done. Patients included IVF couples having day 3 embryos transferred. Rates of clinical pregnancy, twinning, and high order multiples (HOM) were investigated. ResultsWhen one or more HIP embryos were present in the cohort, there were no significant differences in clinical pregnancy rates when the decision was made to reduce the scheduled number of embryos to transfer by 1, that is, when comparing 2 embryos transferred with 2 reduced to 1, 3 with 3 reduced to 2, or 4 with 4 reduced to 3. Twinning rates were the same when at least 2 embryos were transferred and high order multiple rates were not different when at least 3 embryos were transferred. The presence of HIP embryos increases the chance of pregnancy when 1–2 embryos are transferred. With 3–4 embryos, the presence or absence of a HIP embryo becomes less of a factor in establishing pregnancy. ConclusionsSelecting HIP embryos can reduce the number of embryos transferred without adversely affecting pregnancy rates. Their presence is a strong predictor of pregnancy in patients ≤37 years old.

Selecting a Single Embryo for Transfer after In-vitro Fertilization: A Translational Medicine Perspective

Apart from the huge efforts made to develop an embryo in-vitro, the embryologist is faced with an even bigger task to choose an embryo from a cohort that will produce pregnancy. Selecting a single viable one is often a challenge to the embryologists and clinicians involved. This is because the one that is selected may determine outcome of the cycle. It is not in the embryologist interest to choose an unhealthy embryo that will allow patients suffer emotionally or financially. A responsible professional often seeks way to select embryo that will make a baby. There are several methods that have been introduced for this purpose since the birth of Louis Brown in 1978. These methods include the traditional embryo morphology grading system, newer technologies like time lapse monitoring, pre-implantation genetic screening metabolomics, proteomics and transcriptomics. There are conflicting claims that these newer technologies are superior to the morphology grading system. This paper however, advocates for a holistic system that is capable of analyzing all the pathways in an embryo to provide a general health picture. Newer technologies for embryo selection and the traditional embryo grading system in its present form have not achieved 100% accuracy in selecting a single embryo. Research into a holistic system capable of selecting a single embryo will be ideal and better driven by a translational research team that will provide an interdisciplinary and multifrontier approach. We have described here an elementary version of such a system with the hope that it simplicity, rapid turnaround time and cost effectiveness will encourage it application in IVF. This diagnostic platform may be developed for use in IVF laboratory in the future to select the most appropriate embryo.

Grupo de Interés de Embriología de ASEBIR: estudio multicéntrico para la validación del criterio ASEBIR de valoración morfológica de embriones tempranos en día +3 y su asociación con la tasa de nacido vivo

IntroductionIn 2007, Spanish Association for the Study of Reproductive Biology developed a standardized embryo grading system. After 6 years of application, it has been considered necessary to evaluate its validity, not only in terms of implantation rate, as in most studies, but also in terms of live-birth rate. Material and methodsA multicenter retrospective study was conducted on 1,451 embryo cycles performed from January to December 2011, in which 2,678 embryos were transferred on day 3. The inclusion criteria were fresh cycles with autologous oocytes (women age≤35 years old) or donor oocytes, cycles with a 100 or a 0% implantation rate or from homogenous embryo transfers (embryos transferred were of the same grade). Apart from the collection of embryo morphology parameters, additional variables such as maternal age, female sterility etiology, type of treatment (oocyte donation or autologous cycles), insemination technique (standard IVF or ICSI) or the sperm quality on the insemination day were added to the data collection. Statistical analysis: Chi square, Fisher's exact test and binary logistic regression analysis. P<.05 was considered to be statistically significant. ResultsAnalysis of data revealed a positive correlation between embryo quality (grade) and both implantation and live-birth rate, which are maximum in grade A embryos (25.3 and 21.7%, respectively) and minimal in grade D embryos (9.7 and 7.1%, respectively). DiscussionThe current study demonstrates that the Spanish Association for the Study of Reproductive Biology embryo grading system has predictive value on the implantation and live-birth rate for early embryos on day 3. However, this grading system is open to future modifications, based on knowledge and scientific evidence, which could increase the accuracy of successful prediction in our patients.