Should there be an "AI" in TEAM? Embryologists selection of high implantation potential embryos improves with the aid of an artificial intelligence algorithm.

PURPOSE: A deep learning artificial intelligence (AI) algorithm has been demonstrated to outperform embryologists in identifying euploid embryos destined to implant with an accuracy of 75.3% (1). Our aim was to evaluate the performance of highly trained embryologists in selecting top quality day 5 euploid blastocysts with and without the aid of a deep learning algorithm. MATERIALS AND METHODS: A non-overlapping series of 200 sets of day 5 euploid embryo images with known implantation outcomes was distributed to 17 highly trained embryologists. One embryo in each set was known to have implanted and one failed implantation. They were asked to select which embryo to transfer from each set. The same 200 sets of embryos, with indication of which embryo in each set had been identified by the algorithm as more likely to implant was then distributed. Chi-squared, t-test, and receiver operating curves were performed to compare the embryologist performeance with and without AI. RESULTS: Fourteen embryologists completed both assessments. Embryologists provided with AI results selected successfully implanted embryos in 73.6% of cases compared to 65.5% for those selected using visual assessments alone (p < 0.001). All embryologists improved in their ability to select embryos with the aid of the AI algorithm with a mean percent improvement of 11.1% (range 1.4% to 15.5%). There were no differences in degree of improvement by embryologist level of experience (junior, intermediate, senior). CONCLUSIONS: The incorporation of an AI framework for blastocyst selection enhanced the performance of trained embryologists in identifying PGT-A euploid embryos destined to implant.

Using outcome data from one thousand mosaic embryo transfers to formulate an embryo ranking system for clinical use.

OBJECTIVE: To study how the attributes of mosaicism identified during preimplantation genetic testing for aneuploidy relate to clinical outcomes, in order to formulate a ranking system of mosaic embryos for intrauterine transfer. DESIGN: Compiled analysis. SETTING: Multi-center. PATIENT(S): A total of 5,561 euploid blastocysts and 1,000 mosaic blastocysts used in clinical transfers in patients undergoing fertility treatment. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Implantation (gestational sac), ongoing pregnancy, birth, and spontaneous abortion (miscarriage before 20 weeks of gestation). RESULT(S): The euploid group had significantly more favorable rates of implantation and ongoing pregnancy/birth (OP/B) compared with the combined mosaic group or the mosaic group affecting only whole chromosomes (implantation: 57.2% vs. 46.5% vs. 41.8%; OP/B: 52.3% vs. 37.0% vs. 31.3%), as well as lower likelihood of spontaneous abortion (8.6% vs. 20.4% vs. 25%). Whole-chromosome mosaic embryos with level (percent aneuploid cells) <50% had significantly more favorable outcomes than the ≥50% group (implantation: 44.5% vs. 30.4%; OP/B: 36.1% vs. 19.3%). Mosaic type (nature of the aneuploidy implicated in mosaicism) affected outcomes, with a significant correlation between number of affected chromosomes and unfavorable outcomes. This ranged from mosaicism involving segmental abnormalities to complex aneuploidies affecting three or more chromosomes (implantation: 51.6% vs. 30.4%; OP/B: 43.1% vs. 20.8%). Combining mosaic level, type, and embryo morphology revealed the order of subcategories regarding likelihood of positive outcome. CONCLUSION(S): This compiled analysis revealed traits of mosaicism identified with preimplantation genetic testing for aneuploidy that affected outcomes in a statistically significant manner, enabling the formulation of an evidence-based prioritization scheme for mosaic embryos in the clinic.

Assessment of embryo implantation potential with a cloud-based automatic software.

RESEARCH QUESTION: Is embryo selection by Dana (automatic software for embryo evaluation) associated with a higher implantation rate in IVF treatments? DESIGN: A three-phase study for Dana system's validation: creation of a data-cloud of known implantation data (KID) embryos from 1676 transferred embryos; embryo evaluation by Dana considering manual annotations and embryo development videos (389 transferred embryos); and validation of Dana automatic selection, without embryologist's intervention (147 transferred embryos); RESULTS: The implantation rate of the 1021 KID embryos from phase 1 served to set four grades of embryos referring to implantation rate: A = 34%, B = 25%, C = 24%, and D = 19%. Phase 2: a classification ranking according to the unit average distance (UAD) and implantation potential was established: top (UAD ≤0.50), high (UAD = 0.51-0.66), medium (UAD = 0.67-1.03) and low (UAD >1.03). Pregnancy rates were 59%, 46%, 36% and 28%, respectively (P < 0.001). Phase 3: embryos were automatically categorized according to Dana's classification ranking. Most implanted embryos were found in groups top, high and medium (UAD ≤1.03), whereas the implantation rate in group low (UAD >1.03) was significantly lower: 46% versus 25%, respectively (P = 0.037). The twin gestation rate was higher when number of top embryos (UAD ≤0.5) transferred were two (52%) versus one (25%) (P < 0.001). CONCLUSIONS: Embryo selection based on Dana ranking increases the success of IVF treatments at least in oocyte donation programmes. The multicentre nature of the study supports its applicability at different clinics, standardizing the embryo development's interpretation. Dana's innovation is that the system increases its accuracy as the database grows.

Performance of a deep learning based neural network in the selection of human blastocysts for implantation.

Deep learning in in vitro fertilization is currently being evaluated in the development of assistive tools for the determination of transfer order and implantation potential using time-lapse data collected through expensive imaging hardware. Assistive tools and algorithms that can work with static images, however, can help in improving the access to care by enabling their use with images acquired from traditional microscopes that are available to virtually all fertility centers. Here, we evaluated the use of a deep convolutional neural network (CNN), trained using single timepoint images of embryos collected at 113 hr post-insemination, in embryo selection amongst 97 clinical patient cohorts (742 embryos) and observed an accuracy of 90% in choosing the highest quality embryo available. Furthermore, a CNN trained to assess an embryo's implantation potential directly using a set of 97 euploid embryos capable of implantation outperformed 15 trained embryologists (75.26% vs. 67.35%, p<0.0001) from five different fertility centers.

Around one in seven couples have trouble conceiving, which means there is a high demand for solutions such as in vitro fertilization, also known as IVF. This process involves fertilizing and developing embryos in the laboratory and then selecting a few to implant into the womb of the patient. IVF, however, only has a 30% success rate, is expensive and can be both mentally and physically taxing for patients. Selecting the right embryos to implant is therefore extremely important, as this increases the chance of success, minimizes complications and ensures the baby will be healthy. Currently the tools available for making this decision are limited, highly subjective, time-consuming, and often extremely expensive. As a result, embryologists often rely on their experience and observational skills when choosing which embryos to implant, which can lead to a lot of variability. An automated system based on artificial intelligence (AI) could therefore improve IVF success rates by assisting embryologists with this decision and ensuring more consistent results. The AI system could learn how embryos develop over time and then uses this information to select the best embryos to implant from just a single image. This would offer a cheaper alternative to current analysis tools that are only available at the most expensive IVF clinics. Now, Bormann, Kanakasabapathy, Thirumalaraj et al. have developed an AI system for IVF based on thousands of images of embryos. Using individual images, the system selected embryos of a comparable quality to those selected by a human specialist. It also showed a greater ability to identify embryos that will lead to successful implantation. Indeed, the software outperformed 15 embryologists from five different centers across the United States in detecting which embryos were most likely to implant out of a group of high-quality embryos with few visible differences. Artificial intelligence has many potential applications to support expert clinical decision-making. Systems like these could improve success, reduce errors and lead to faster, cheaper and more accessible results. Beyond immediate IVF applications, this system could also be used in research and industry to help understand differences in embryo quality.

Clinical and pregnancy outcomes of double and single blastocyst transfers related with morphological grades in vitrified-warmed embryo transfer.

OBJECTIVE: To evaluate clinical and pregnancy outcomes of double and single blastocyst transfers related with morphological grades in vitrified-warmed embryo transfer. MATERIALS AND METHODS: In a retrospective cohort analysis, data were assessed from women who underwent vitrified-warmed blastocyst transfers (VBT) at CHA Gangnam Medical Center between 2014 and 2015. All VBT cycles were categorized into three groups according to blastocyst quality: GG (double good blastocysts transfer), GP (one good and one poor blastocyst transfer), and GS (single good blastocyst transfer). Blastocysts were graded morphologically and ⩾3BB grade was considered good quality. RESULTS: There were 628 transfers in group GG, 401 transfers in group GP, and 277 transfers in group GS. Both clinical pregnancy rate (CPR) and live birth rate (LBR) were the highest in group GG (CPR 65.9%, LBR 55.3%, p < 0.001), but not significantly different between group GP and GS. Multiple pregnancy rates increased significantly in the following order: GS (1.4%), GP (13.5%), and GG (25.6%). Single LBR was the highest in group GS (38.6%, p < 0.001). CONCLUSION: As an effective VBT, especially for reducing multiple pregnancy and increasing single live birth, single good blastocyst transfer may be recommended rather than any double blastocyst transfer methods. Moreover, transferring a good and a poor blastocyst simultaneously should be avoided.

Biodiversity of protists and nematodes in the wild nonhuman primate gut.

Documenting the natural diversity of eukaryotic organisms in the nonhuman primate (NHP) gut is important for understanding the evolution of the mammalian gut microbiome, its role in digestion, health and disease, and the consequences of anthropogenic change on primate biology and conservation. Despite the ecological significance of gut-associated eukaryotes, little is known about the factors that influence their assembly and diversity in mammals. In this study, we used an 18S rRNA gene fragment metabarcoding approach to assess the eukaryotic assemblage of 62 individuals representing 16 NHP species. We find that cercopithecoids, and especially the cercopithecines, have substantially higher alpha diversity than other NHP groups. Gut-associated protists and nematodes are widespread among NHPs, consistent with their ancient association with NHP hosts. However, we do not find a consistent signal of phylosymbiosis or host-species specificity. Rather, gut eukaryotes are only weakly structured by primate phylogeny with minimal signal from diet, in contrast to previous reports of NHP gut bacteria. The results of this study indicate that gut-associated eukaryotes offer different information than gut-associated bacteria and add to our understanding of the structure of the gut microbiome.

Developmental and cytogenetic assessments of preimplantation embryos derived from in-vivo or in-vitro matured human oocytes.

Aneuploidy is of great relevance to embryo selection, as it represents one of the important causes of implantation failure. Furthermore, immature oocytes, retrieved during gonadotrophin-stimulated IVF cycles, are generally discarded in clinics; whereas, there was no detectable comprehensive evidence on higher rates of aneuploidy based on maturity status on the day of oocyte retrieval. As well, the correlation between embryo morphology on aneuploidy remains unclear. The aim was to evaluate the developmental and genetic integrity of human preimplantation embryos from rescue in-vitro matured MII stage oocytes as well as in vivo matured oocytes. 541 rescue in-vitro matured oocytes as case as well as 659 in-vivo matured oocytes as control were used for the developmental assay. Finally, 121 cleaved embryos with good quality were analyzed by FISH technique for the detection of chromosomes X, Y, 13, 15, 16, 18, 21 and 22. The fertilization rates were 61.62% and 61.76% in case and control groups, respectively. Also, embryo formation rates of 89.1% vs. 92.2% were recorded for case and control groups, respectively. Good quality embryos on day 3 were 62.54% in case and 68.36% in control groups. There were insignificant differences in fertilization, embryo formation and quality between the groups. Total abnormality in 35 of the 60 embryos was 58.5% in case and 62.3% in control (p < 0.05). There were significant differences between aneuploidy rates of embryos using only sex chromosome preimplantation genetic screening (PGS) and sex chromosome in combination with autosomal chromosomes PGS in case (58.5% vs 28.3%, p = 0.000) and control groups (62.3% vs 21.3%; p = 0.000). The results demonstrated that a high proportion of good quality embryos were aneuploid in both patient groups with no obvious increase in aneuploidies as a result of rescue IVM application. Furthermore, the morphological characteristics of embryos do not completely consistent with chromosomal content. Despite the Rescue IVM is currently not a routine procedure in association with IVF, our finding suggested a viable option for young infertile women facing cancellation of their IVF treatment due to ovarian over-response or resistance factors as well as patients with low functional ovarian reserve considering good quality of embryos from rescue IVM-MII oocytes.

Automatic Identification of Human Blastocyst Components via Texture.

Choosing the most viable embryo during human in vitro fertilization (IVF) is a prime factor in maximizing pregnancy rate. Embryologists visually inspect morphological structures of blastocysts under microscopes to gauge their health. Such grading introduces subjectivity amongst embryologists and adds to the difficulty of quality control during IVF. In this paper, we introduce an algorithm for automatic segmentation of two main components of human blastocysts named: Trophectoderm (TE) and inner cell mass (ICM). We utilize texture information along with biological and physical characteristics of day-5 human embryos (blastocysts) to identify TE or ICM regions according to their intrinsic properties. Both these regions are highly textured and very similar in the quality of their texture, and they often look connected to each other when imaged. These attributes make their automatic identification and separation from each other a difficult task even for an expert embryologist. By automatically identifying TE and ICM regions, we offer the opportunity to perform more detailed assessment of blastocysts. This could help in analyzing, in a quantitative way, various visual/geometrical characteristics of these regions that when combined with the pregnancy outcome can determine the predictive values of such attributes. Our work aids future research in understanding why certain embryos have higher pregnancy success rates. This paper is tested on a set of 211 blastocyst images. We report an accuracy of 86.6% for identification of TE and 91.3% for ICM.

Selection of Suitable Internal Control Genes for Accurate Normalization of Real-Time Quantitative PCR Data of Buffalo (Bubalus bubalis) Blastocysts Produced by SCNT and IVF.

We evaluated the suitability of 10 candidate internal control genes (ICGs), belonging to different functional classes, namely ACTB, EEF1A1, GAPDH, HPRT1, HMBS, RPS15, RPS18, RPS23, SDHA, and UBC for normalizing the real-time quantitative polymerase chain reaction (qPCR) data of blastocyst-stage buffalo embryos produced by hand-made cloning and in vitro fertilization (IVF). Total RNA was isolated from three pools, each of cloned and IVF blastocysts (n = 50/pool) for cDNA synthesis. Two different statistical algorithms geNorm and NormFinder were used for evaluating the stability of these genes. Based on gene stability measure (M value) and pairwise variation (V value), calculated by geNorm analysis, the most stable ICGs were RPS15, HPRT1, and ACTB for cloned blastocysts, HMBS, UBC, and HPRT1 for IVF blastocysts and RPS15, GAPDH, and HPRT1 for both the embryo types analyzed together. RPS18 was the least stable gene for both cloned and IVF blastocysts. Following NormFinder analysis, the order of stability was RPS15 = HPRT1>GAPDH for cloned blastocysts, HMBS = UBC>RPS23 for IVF blastocysts, and HPRT1>GAPDH>RPS15 for cloned and IVF blastocysts together. These results suggest that despite overlapping of the three most stable ICGs between cloned and IVF blastocysts, the panel of ICGs selected for normalization of qPCR data of cloned and IVF blastocyst-stage embryos should be different.

Blastocyst classification systems used in Latin America: is a consensus possible?

OBJECTIVE: To identify different blastocyst classification systems used by embryologists in Latin American countries and evaluate the possibility of establishing a consensus among these countries. METHODS: An E-mail survey was carried out through the Latin American Network of Assisted Reproduction (REDLARA) aimed at embryologists from assisted reproduction centers in Latin countries. RESULTS: Sixty surveys were collected from 12 Latin American countries, of which 66.7% had >10years of professional practice as embryologists. Seven different blastocyst classification systems were reported, of which 5 have previously been described in the literature. CONCLUSION: Although the group of embryologists surveyed use different blastocyst classification systems, most in this group consider that the embryo score system should be unified in their countries as well as in the region.

Human Embryo Morphokinetics Correlation with Obstetric Outcomes and Antepartum Biomarkers.

OBJECTIVE: To evaluate the correlation between em-bryo morphology and antepartum biomarker levels and obstetric outcomes following single embryo transfers (SETs). STUDY DESIGN: In vitro fertilization patients who delivered at the study site's partnering hospital were included. Embryo morphology was compared to antepartum biomarkers and pregnancy outcomes. RESULTS: At a single academic reproductive endocrinology/infertility center, 2,882 SETs (day 3, 5, or 6) were performed from July 1, 2001, to June 30, 2013. Embryologic, serologic, and obstetric outcomes were available for 78 patients (cleavage [n=23]; blastocyst [n=55]). Embryos with a lower quality inner cell mass (ICM) demonstrated higher second trimester maternal serum alpha-fetoprotein (AFP) (p=0.0241) and AFP multiple of median (MoM) (p= 0.0036). After adjusting for age, body mass index, parity, and smoking status, this statistical significance remained (p=0.0319 for AFP; p=0.0026 for AFP MoM). CONCLUSION: Embryo morphokinetics did not correlate with perinatal outcome or a majority of placental biomarkers. A low ICM score was associated with higher second trimester maternal serum AFP values. This warrants further investigation, especially due to the risks of potential adverse obstetric outcomes associated with high AFP levels. Despite these findings, no difference in obstetric outcomes between low and high ICM scores was observed after controlling for maternal confounding variables. Patients should be reassured that embryo implantation, regardless of morphological score, results in similar obstetric outcomes.

The impact of age on blastocyst scoring after single and double embryo transfers.

OBJECTIVE: The aim of the present study was to assess the putative relationship between patient's age and blastocyst scores, in single (SET) or double (DET) transfer cycles, that resulted in single or twin pregnancy or non-pregnancy. Second, we analyzed the effect of maternal age on clinical gestation and implantation rates after single and double blastocyst transfers. METHODS: Retrospective analysis of 164 assisted reproduction cycles with embryo transfers. RESULTS: Data demonstrated that for both, young (<35 years of age) and older (≥35 years of age) women, trophectoderm (TE) score is the most important parameter to assess concerning embryo selection. However, inner cell mass (ICM) also plays an important role on blastocyst selection in the group of older patients. In addition, our data shows that for young women the transfer of a single blastocyst results in similar gestational rates as those for DET. CONCLUSION: We suggest that blastocyst grading for patients aged 35 years or above shall be performed using a strict grading policy, possibly not of a single parameter, but TE, ICM and expansion grades together, to choose the "best combined-score blastocyst". DETs should be considered, particularly after previous cycles with pregnancy failures.

Elective Single Embryo Transfer: an update to UK Best Practice Guidelines.

A significant number of multiple pregnancies and births worldwide continue to occur following treatment with Assisted Reproductive Technologies (ARTs). Whilst efforts have been made to increase the proportion of elective single embryo transfer (eSET) cycles, the multiple pregnancy rate or MPR remains at a level that most consider unacceptable given the associated clinical risks to mothers and babies, and the additional costs associated with neonatal care of premature and low birth weight babies. Northern Europe, Australia and Japan have continued to lead the way in the adoption of eSET. Randomised controlled trials or RCTs, meta-analyses and economic analyses support the implementation of an eSET policy, particularly in light of recent advances in ARTs. This paper provides a review of current evidence and an update to the eSET guidelines first published by Cutting et al. (2008) intended to assist ART clinics in the implementation of an effective eSET policy.

A clinically useful simplified blastocyst grading system.

The aim of this study was to investigate whether a new simplified blastocyst grading system (A: fully expanded, clear inner cell mass, cohesive trophectoderm; B: not yet expanded, clear inner cell mass, cohesive trophectoderm; C: small inner cell mass ± irregular trophectoderm ± excluded/degenerate cells) was clinically useful. All day-5 single embryo transfers between 15 June 2009 and 29 June 2012 were reviewed. Implantation, clinical pregnancy and live birth rates were related to embryo quality. Five embryologists were asked to grade and decide the clinical fate of 80 images of day-5 embryos on two occasions 4-6 weeks apart. Implantation, clinical pregnancy and live birth rates decreased with deteriorating embryo quality. A highly significant (P < 0.01) difference was observed between the groups. Inter-observer agreement was substantial for grade allocation (K = 0.63) and clinical decision-making (K = 0.66). Intra-observer agreement ranged from substantial (K = 0.71) to almost perfect (K = 0.88) for grade allocation, and was almost perfect for clinical fate determination (K ≥ 0.84). This grading system is quick and easy to use, effectively predicts IVF outcome and has levels of agreement similar to, if not better than, those associated with more complex grading systems.

Analysis of embryo morphokinetics, multinucleation and cleavage anomalies using continuous time-lapse monitoring in blastocyst transfer cycles.

BACKGROUND: Time-lapse imaging combined with embryo morphokinetics may offer a non-invasive means for improving embryo selection. Data from clinics worldwide are necessary to compare and ultimately develop embryo classifications models using kinetic data. The primary objective of this study was to determine if there were kinetic differences between embryos with limited potential and those more often associated with in vitro blastocyst formation and/or implantation. We also wanted to compare putative kinetic markers for embryo selection as proposed by other laboratories to what we were observing in our own laboratory setting. METHODS: Kinetic data and cycle outcomes were retrospectively analyzed in patients age 39 and younger with 7 or more zygotes cultured in the Embryoscope. Timing of specific events from the point of insemination were determined using time-lapse (TL) imaging. The following kinetic markers were assessed: time to syngamy (tPNf), t2, time to two cells (c), 3c (t3), 4c ( t4), 5c (t5), 8c (t8), morula (tMor), start of blastulation (tSB); tBL, blastocyst (tBL); expanded blastocyst (tEBL). Durations of the second (cc2) and third (cc3) cell cycles, the t5-t2 interval as well as time to complete synchronous divisions s1, s2 and s3 were calculated. Incidence and impact on development of nuclear and cleavage anomalies were also assessed. RESULTS: A total of 648 embryos transferred on day 5 were analyzed. The clinical pregnancy and implantation rate were 72% and 50%, respectively. Morphokinetic data showed that tPNf, t2,t4, t8, s1, s2,s3 and cc2 were significantly different in embryos forming blastocysts (ET or frozen) versus those with limited potential either failing to blastulate or else forming poor quality blastocysts ,ultimately discarded. Comparison of embryo kinetics in cycles with all embryos implanting (KID+) versus no implantation (KID-) suggested that markers of embryo competence to implant may be different from ability to form a blastocyst. The incidence of multinucleation and reverse cleavage amongst the embryos observed was 25% and 7%, respectively. Over 40% of embryos exhibiting these characteristics did however form blastocysts meeting our criteria for freezing. CONCLUSIONS: These data provide us with a platform with which to potentially enhance embryo selection for transfer.

Morphological systems of human embryo assessment and clinical evidence.

Success rates with IVF have improved remarkably since the procedure was first established for clinical use with the first successful birth in 1978. The main goals today are to perform single-embryo transfer in order to prevent multiple pregnancies and achieve higher overall pregnancy rates. However, the ability to identify the most viable embryo in a cohort remains a challenge despite the numerous scoring systems currently in use. Clinicians still depend on developmental rate and morphological assessment using light microscopy as the first-line approach for embryo selection. Active research in the field involves developing non-invasive methods for scoring embryos and ranking them according to their ability to implant and give rise to a healthy birth. Current attention is particularly being focused on time-lapse evaluation. Available data from preliminary studies indicate that these systems are safe;prospective data now need to be collected to determine whether these methods do improve implantation rates. This review gives brief consideration to the use of morphological evaluations in assisted reproduction treatment, discusses the types of embryo scoring,digital imaging and biometric approaches currently in use and comments on future developments for embryo evaluation.

Artificial intelligence techniques for embryo and oocyte classification.

One of the most relevant aspects in assisted reproduction technology is the possibility of characterizing and identifying the most viable oocytes or embryos. In most cases, embryologists select them by visual examination and their evaluation is totally subjective. Recently, due to the rapid growth in the capacity to extract texture descriptors from a given image, a growing interest has been shown in the use of artificial intelligence methods for embryo or oocyte scoring/selection in IVF programmes. This work concentrates the efforts on the possible prediction of the quality of embryos and oocytes in order to improve the performance of assisted reproduction technology, starting from their images. The artificial intelligence system proposed in this work is based on a set of Levenberg-Marquardt neural networks trained using textural descriptors (the local binary patterns). The proposed system was tested on two data sets of 269 oocytes and 269 corresponding embryos from 104 women and compared with other machine learning methods already proposed in the past for similar classification problems. Although the results are only preliminary, they show an interesting classification performance. This technique may be of particular interest in those countries where legislation restricts embryo selection. One of the most relevant aspects in assisted reproduction technology is the possibility of characterizing and identifying the most viable oocytes or embryos. In most cases, embryologists select them by visual examination and their evaluation is totally subjective. Recently, due to the rapid growth in our capacity to extract texture descriptors from a given image, a growing interest has been shown in the use of artificial intelligence methods for embryo or oocyte scoring/selection in IVF programmes. In this work, we concentrate our efforts on the possible prediction of the quality of embryos and oocytes in order to improve the performance of assisted reproduction technology, starting from their images. The artificial intelligence system proposed in this work is based on a set of Levenberg-Marquardt neural networks trained using textural descriptors (the 'local binary patterns'). The proposed system is tested on two data sets, of 269 oocytes and 269 corresponding embryos from 104 women, and compared with other machine learning methods already proposed in the past for similar classification problems. Although the results are only preliminary, they showed an interesting classification performance. This technique may be of particular interest in those countries where legislation restricts embryo selection.

[Non invasive evaluation of the embryo: morphology of preimplantation embryos]. / Evaluation non invasive de l'embryon : morphologie embryonnaire préimplantatoire.

From all the possibilities of assessment of the embryonic viability in vitro, morphology observed at microscope is the most elective criteria because of its specificity, sensibility and its armless side. The use of classifications at each stage of the development is essential for the selection of the right embryo to be replaced but do not allow dynamic evaluation. The development of computerized tools to capture and analyse images must emancipate us from oversimplification that doesn't cover frequent daily situations. The purpose is to examine the most frequent widespread classifications, the proposed alternatives and the consequences on embryonic replacement policy. More than a simple morphological factual evaluation of an attempt, it would be interesting to correlate embryonic development to stimulation criteria or quality culture media and other variables to enhance the obtaining of good prognostic embryos. The retrospective analysis of the databases of images will probably respond to this preoccupation.

Classification of morphological changes based on the number of cleavage divisions in bovine embryos.

Quantification based on cleavage division (CD) of bovine preimplantation embryos facilitates quantitative analyses of embryonic developmental processes because CD occurs roughly once each day for all blastomeres for up to at least 9 days after ovulation. Therefore, embryonic morphological changes during this period were classified according to CD number. In this study, embryos collected from superovulated donors 0-9 days after ovulation were first classified morphologically into 14 conventional developmental stages. The total cell numbers (TCN) of embryos were measured using the air-dry method. The respective CD numbers of the embryos were then determined using logarithmic transformation of the TCN. The CD numbers of embryos were increased 0-10th with 11 stages. The 0th CD corresponded to 1-cell stage embryos; the 1st CD corresponded to 2-cell stage embryos; the 2nd CD corresponded to 3-4-cell stage embryos; the 3rd CD corresponded to 5-8-cell stage embryos; the 4th CD corresponded to 9-16-cell stage embryos, the 5th CD corresponded to morulae (17-32-cell stage embryos); and the 6th CD corresponded to the compact morulae. Furthermore, the 7th CD included early blastocysts to blastocysts. The 8th CD included expanded, collapsed and hatching blastocysts. The 9th CD included hatched blastocysts. The 10th CD included expanding-hatched blastocysts. The relationship between the CD number and the morphological characteristics of the bovine embryos 0-9 days after ovulation was expressed using a linear equation, and this revealed a high degree of correlation (y=0.98x-0.96, r=0.99). These results suggest that morphological changes of bovine embryos can be classified accurately using an 11-stage classification system based on the number of cleavages.

Blastocyst production from in vitro-produced day-2 bovine embryos classified by cleavage stage, and cytogenetical evaluation of the resultant day-8 blastocysts.

The present study was conducted to determine the criteria for selecting good quality embryos on Day-2 post-insemination and at the blastocyst stage. Bovine oocytes were matured, fertilized and cultured in vitro. First, Day-2 embryos were classified based on the number of blastomeres into 2-cell, 3- to 4-cell, 5- to 8-cell and >8-cell stage embryos; chromosome samples were then prepared. In the second experiment, the Day-2 embryos classified according to the number of blastomeres were cultured separately for an additional 6 days (Day 8). The resultant Day-8 blastocysts from each group of Day-2 embryos were classified into the following 3 grades based on morphology and diameter: Grade A, hatched and hatching blastocysts; Grade B, expanded blastocysts; and Grade C, unexpanded blastocysts. Chromosome samples were then prepared. The 5- to 8-cell stage Day-2 embryos had the lowest incidence of chromosomal abnormalities (13.5%, P<0.05) and the highest development rate to blastocysts (59.2%, P<0.05). Furthermore, the blastocysts derived from the 5- to 8-cell stage embryos had the largest mean number of cells (102.8+/-42.4, P<0.05), largest number of metaphases per blastocyst (9.5+/-4.8, P<0.05) and lowest incidence of chromosomal abnormalities (24.6%, P<0.05). The Grade A blastocysts had the largest mean number of cells (136.6+/-33.4, P<0.05), a large number of metaphases per blastocyst (11.9+/-5.5, P<0.05) and a low incidence of severe chromosomal abnormalities (17.3%). The results showed that, at Day 2, the 5- to 8-cell stage embryos were of better quality since they had the lowest incidence of chromosomal abnormalities and the highest blastocyst rate and the resultant blastocysts had the largest number of cells and lowest incidence of chromosomal abnormalities. In particular, selection of Grade A blastocysts can improve the development rate to term.