A review on automatic analysis of human embryo microscope images

EVATOM: an optical, label-free, machine learning assisted embryo health assessment tool

The combination of a good quality embryo and proper maternal health factors promise higher chances of a successful in vitro fertilization (IVF) procedure leading to clinical pregnancy and live birth. Of these two factors, selection of a good embryo is a controllable aspect. The current gold standard in clinical practice is visual assessment of an embryo based on its morphological appearance by trained embryologists. More recently, machine learning has been incorporated into embryo selection "packages". Here, we report EVATOM: a machine-learning assisted embryo health assessment tool utilizing an optical quantitative phase imaging technique called artificial confocal microscopy (ACM). We present a label-free nucleus detection method with, to the best of our knowledge, novel quantitative embryo health biomarkers. Two viability assessment models are presented for grading embryos into two classes: healthy/intermediate (H/I) or sick (S) class. The models achieve a weighted F1 score of 1.0 and 0.99 respectively on the in-distribution test set of 72 fixed embryos and a weighted F1 score of 0.9 and 0.95 respectively on the out-of-distribution test dataset of 19 time-instances from 8 live embryos.

Novel image analyser-assisted morphometric methodology offer unique opportunity for selection of embryos with potential for implantation

BACKGROUND

Previous studies looked into the connections between pregnancy and the Zona Pellucida (ZP) thickness and Zona Pellucida Thickness Variation (ZPTV), as well as the embryo's radius, circumference, perimeter and global symmetry. However, no research has linked embryo implantation and pregnancy to the percentage of ZP thinning, the reduction in ooplasm volume, and the increase in perivitelline space (PVS) volume. Our objective is to correlate the percentage of ZP thinning, the percentage of ooplasm volume shrinkage and the percentage of PVS increase to the implantation. These data will be used for embryo selection as well as it can be put into a software that will assist embryo selection.

MATERIALS AND METHODS

Retrospective study included 281 patients, all of them had 2 embryos transferred, 149 patients got pregnant with two gestation sacs and 132 patients did not get pregnant. All of the transferred embryos had the ZP thickness measured several times from time of ICSI till Embryo Transfer (ET), the ooplasm volume was calculated from time of ICSI till two Pronuclei (2PN) fading and the PVS was calculated from the ICSI time till the 2PN fading.

RESULTS

The first characteristic is the change in the average ZP thickness that decreased by 32.7% + 5.3% at 70 h for the implanted embryos (Group 1) versus 23.6% + 4.8% for non-implanted embryos (Group 2) p = 0.000. The second characteristic is the average reduction in the volume of the ooplasm which is 20.5% + 4.3% in Group 1 versus 15.1% + 5.2% in Group 2, p = 0.000. The third characteristic is the increase in the volume of the PVS which was 38.1% + 7.6% in Group 1 versus 31.6% + 9.7% in Group 2 p = 0.000.

CONCLUSION

The implanted embryos showed higher percent of ZP thinning, higher percent of ooplasm reduction and higher percent of PVS increase.

Using Unlabeled Information of Embryo Siblings from the Same Cohort Cycle to Enhance In Vitro Fertilization Implantation Prediction

High-content time-lapse embryo imaging assessed by machine learning is revolutionizing the field of in vitro fertilization (IVF). However, the vast majority of IVF embryos are not transferred to the uterus, and these masses of embryos with unknown implantation outcomes are ignored in current efforts that aim to predict implantation. Here, whether, and to what extent the information encoded within "sibling" embryos from the same IVF cohort contributes to the performance of machine learning-based implantation prediction is explored. First, it is shown that the implantation outcome is correlated with attributes derived from the cohort siblings. Second, it is demonstrated that this unlabeled data boosts implantation prediction performance. Third, the cohort properties driving embryo prediction, especially those that rescued erroneous predictions, are characterized. The results suggest that predictive models for embryo implantation can benefit from the overlooked, widely available unlabeled data of sibling embryos by reducing the inherent noise of the individual transferred embryo.

Machine learning assisted health viability assay for mouse embryos with artificial confocal microscopy (ACM)

The combination of a good quality embryo and proper maternal health factors promise higher chances of a successful in vitro fertilization (IVF) procedure leading to clinical pregnancy and live birth. Of these two factors, selection of a good embryo is a controllable aspect. The current gold standard in clinical practice is visual assessment of an embryo based on its morphological appearance by trained embryologists. More recently, machine learning has been incorporated into embryo selection "packages". Here, we report a machine-learning assisted embryo health assessment tool utilizing a quantitative phase imaging technique called artificial confocal microscopy (ACM). We present a label-free nucleus detection method with novel quantitative embryo health biomarkers. Two viability assessment models are presented for grading embryos into two classes: healthy/intermediate (H/I) or sick (S) class. The models achieve a weighted F1 score of 1.0 and 0.99 respectively on the in-distribution test set of 72 fixed embryos and a weighted F1 score of 0.9 and 0.95 respectively on the out-of-distribution test dataset of 19 time-instances from 8 live embryos.

LWMA-Net: Light-weighted morphology attention learning for human embryo grading

Visual inspection of embryo morphology is routinely used in embryo assessment and selection. However, due to the complexity of morphologies and large inter- and intra-observer variances among embryologists, manual evaluations remain to be subjective and time-consuming. Thus, we proposed a light-weighted morphology attention learning network (LWMA-Net) for automatic assistance on embryo grading. The LWMA-Net integrated a morphology attention module (MAM) to seek the informative features and their locations and a multiscale fusion module (MFM) to increase the features flowing in the model. The LWMA-Net was trained with a primary set of 3599 embryos from 2318 couples that were clinically enrolled between Sep. 2016 and Dec. 2018, and generated area under the receiver operating characteristic curves (AUCs) of 96.88% and 97.58% on 4- and 3-category gradings, respectively. An independent test set comprises 691 embryos from 321 couples between Jan. 2019 and Jan. 2021 were used to test the assisted fertility values on the embryo grading. Five experienced embryologists were invited to regrade the embryos in the independent set with and without the aid of the LWMA-Net three months apart. Embryologists aided by our LWMA-Net significantly improved their grading capabilities with average AUCs improved by 4.98%-5.32% on 4- and 3-category grading tasks, respectively, which suggests good potential of our LWMA-Net on assisted human reproduction.

The effect of chemical treatment of the sheep embryo zona pellucida on the ability of blastocysts to hatch after vitrification and warming

BACKGROUND

The embryo release from the zona pellucida is of prerequisites of successful implantation.

OBJECTIVES

Regarding the negative impact of embryo cryopreservation on the blastocysts hatchability, the aim of the present study was to investigate the effects of treating embryonic zona pellucida with pronase or acidic Tyrode's solution (ATS) before morula formation on the viability, freezability, and hatchability of vitrified-warmed resulted blastocysts.

METHODS

In the first experiment, the zona pellucida of 3- and 4-day-old embryos were treated with the above compounds for 30 or 45 s. Then, the competency of the treated embryos to reach to blastocyst stage and the hatchability of resulting blastocysts were investigated. In the second experiment, the cryo-survivability and hatching rate of blastocysts resulting from 3-day-old embryos treated with pronase and ATS for 30 s were tested.

RESULTS

In the first experiment and in contrast to the 45 s exposure, 30-s exposure of embryos to pronase or ATS did not have negative effect on the viability and development of embryos to blastocyst stage. In the second experiment, the freezability of blastocysts derived from 3-day-old embryos treated with pronase and ATS for 30 s was not different from that of the control group. However, the hatching rate of the pronase group was significantly higher than that of the control group.

CONCLUSION

The results of the present study showed that reducing the thickness of zona pellucida of sheep embryos with pronase had no negative effect on the developmental competency and freezability of the treated embryos and improved the hatchability of vitrified-warmed blastocysts.

Review of computer vision application in in vitro fertilization: the application of deep learning-based computer vision technology in the world of IVF

In vitro fertilization has been regarded as a forefront solution in treating infertility for over four decades, yet its effectiveness has remained relatively low. This could be attributed to the lack of advancements for the method of observing and selecting the most viable embryos for implantation. The conventional morphological assessment of embryos exhibits inevitable drawbacks which include time- and effort-consuming, and imminent risks of bias associated with subjective assessments performed by individual embryologists. A combination of these disadvantages, undeterred by the introduction of the time-lapse incubator technology, has been considered as a prominent contributor to the less preferable success rate of IVF cycles. Nonetheless, a recent surge of AI-based solutions for tasks automation in IVF has been observed. An AI-powered assistant could improve the efficiency of performing certain tasks in addition to offering accurate algorithms that behave as baselines to minimize the subjectivity of the decision-making process. Through a comprehensive review, we have discovered multiple approaches of implementing deep learning technology, each with varying degrees of success, for constructing the automated systems in IVF which could evaluate and even annotate the developmental stages of an embryo.

Evaluation of deep convolutional neural networks in classifying human embryo images based on their morphological quality

A critical factor that influences the success of an in-vitro fertilization (IVF) treatment cycle is the quality of the transferred embryo. Embryo morphology assessments, conventionally performed through manual microscopic analysis suffer from disparities in practice, selection criteria, and subjectivity due to the experience of the embryologist. Convolutional neural networks (CNNs) are powerful, promising algorithms with significant potential for accurate classifications across many object categories. Network architectures and hyper-parameters affect the efficiency of CNNs for any given task. Here, we evaluate multi-layered CNNs developed from scratch and popular deep-learning architectures such as Inception v3, ResNET-50, Inception-ResNET-v2, NASNetLarge, ResNeXt-101, ResNeXt-50, and Xception in differentiating between embryos based on their morphological quality at 113 h post insemination (hpi). Xception performed the best in differentiating between the embryos based on their morphological quality.

Clinical pregnancy is significantly associated with the blastocyst width and area: a time-lapse study

In order to maintain pregnancy rates following single embryo transfer, optimisation of embryo culture and selection is vital. Time-lapse monitoring (TLM) has the potential to play a crucial role by providing sequential images of embryo development and minimal disturbance. Therefore, in this study morphometric assessment of blastocyst area and maximum width was performed in order to evaluate if these parameters are associated with pregnancy outcomes in IVF/ICSI cycles. This is a retrospective study of 664 patients who had elective single blastocyst transfer (eSBT). The EmbryoScope drawing tools were used to measure specific variables such as the maximum blastocyst width and blastocyst area. Our results show that women who were pregnant had significantly (P < 0.01) larger blastocyst width [median (range) μm] 184 (125-239) versus non-pregnant, 160 (120-230)] and area [median (range) μm²] 26099 (12101-45,280) versus non-pregnant women, 22,251 (10992-37,931)]. A univariate logistic regression performed showed that blastocyst width [(OR = 1.026, 95% CI = (1.019, 1.033)] was significant (P < 0.01) and for every μm increase of blastocyst width, the odds of clinical pregnancy increase by 2.6%. A univariate logistic regression performed showed that blastocyst area [(OR = 1.00008, 95% CI = (1.00006, 1.00011)] was significant with P < 0.01. For every μm² increase of blastocyst area, our data showed the odds of clinical pregnancy increase by 0.008%. Hosmer-Lemeshow tests of calibrations were performed to verify calibration. Although our findings show a clear correlation between blastocyst dimensions and the clinical pregnancy rate, further studies are necessary to confirm these observations.

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.

Improvement instead of stability in embryo quality between day 3-5: A possible extra predictor for blastocyst selection

OBJECTIVE

The aim of this study was to evaluate the predictive value of the dynamic morphological development process between cleavage-stage and blastocyst-stage embryos.

STUDY DESIGN

A retrospective study was executed between 2015 and 2017 at Ghent University Hospital. A total of 996 first fresh IVF/ICSI cycles resulting in a single embryo transfer on day 5 were included. Embryos were scored on day 3 and day 5 as excellent, good, moderate or poor based on Alpha/ESHRE guidelines and Gardner and Schoolcraft scoring-system. If embryos changed category between day 3 and 5, the number of steps (between excellent; good; moderate; poor) in positive and negative direction was expressed.

RESULTS

On day 5, the ongoing pregnancy rate (OPR) of excellent embryos was 37.4 %. Univariate analyses showed that on day 5, both a higher cell stage, better inner cell mass and better trophectoderm were significantly associated with an ongoing pregnancy. In case of deterioration in quality of individual embryos between day 3 and day 5, the OPR was significantly lower. Conversely, improvement of embryo quality between day 3 and day 5 showed higher ongoing pregnancy rates (overall OPR of good day-3 embryos improving to excellent day-5 embryos: 30 %; moderate day 3 to excellent day 5: 50 %; poor day 3 to excellent day 5: 42 %; poor day 3 to good day 5: 20 %; poor day 3 to moderate day 5: 16 %). When embryos improved from poor on day 3 to excellent day 5 the OPR was significantly higher in comparison with embryos that did not change in quality scoring during development (steady embryos) (OR: 1.785, p < 0.05).

CONCLUSION

Our results suggest that it is more likely to achieve an ongoing pregnancy when transferring an embryo that has improved in quality between days 3 and 5 as opposed to one that has remained stable.

Consistency and objectivity of automated embryo assessments using deep neural networks

OBJECTIVE

To evaluate the consistency and objectivity of deep neural networks in embryo scoring and making disposition decisions for biopsy and cryopreservation in comparison to grading by highly trained embryologists.

DESIGN

Prospective double-blind study using retrospective data.

SETTING

U.S.-based large academic fertility center.

PATIENTS

Not applicable.

INTERVENTION(S)

Embryo images (748 recorded at 70 hours postinsemination [hpi]) and 742 at 113 hpi) were used to evaluate embryologists and neural networks in embryo grading. The performance of 10 embryologists and a neural network were also evaluated in disposition decision making using 56 embryos.

MAIN OUTCOME MEASURES

Coefficients of variation (%CV) and measures of consistencies were compared.

RESULTS

Embryologists exhibited a high degree of variability (%CV averages: 82.84% for 70 hpi and 44.98% for 113 hpi) in grading embryo. When selecting blastocysts for biopsy or cryopreservation, embryologists had an average consistency of 52.14% and 57.68%, respectively. The neural network outperformed the embryologists in selecting blastocysts for biopsy and cryopreservation with a consistency of 83.92%. Cronbach's α analysis revealed an α coefficient of 0.60 for the embryologists and 1.00 for the network.

CONCLUSIONS

The results of our study show a high degree of interembryologist and intraembryologist variability in scoring embryos, likely due to the subjective nature of traditional morphology grading. This may ultimately lead to less precise disposition decisions and discarding of viable embryos. The application of a deep neural network, as shown in our study, can introduce improved reliability and high consistency during the process of embryo selection and disposition, potentially improving outcomes in an embryology laboratory.

Predicting Human Embryos' Implantation Outcome from a Single Blastocyst Image

Only one-third of embryo transfer cycles via invitro fertilization, the most common fertility treatment, leads to a clinical pregnancy. Identifying embryos with the highest potentials for transfer is an essential step to optimize in-vitro fertilization outcome. However, human embryos are complicated by nature and some of their developmental aspects has still remained a mystery to expert biologists. In this paper, the first-ever attempt is made to estimate probability of implantation using a single blastocyst image. First, a semantic segmentation system is proposed for human blastocyst components in microscopic images. Second, a multi-stream classification model is proposed for the prediction of embryos' implantation outcome. The proposed classification model features an architectural component, Compact-Contextualize-Calibrate (C³) to guide the feature extraction process and a slow-fusion strategy to learn cross-modality features. Experimental results confirm that the proposed method delivers the first-reported implantation outcome prediction via a single blastocyst image to date with a mean accuracy of 70.9%.

Cell extraction automation in single cell surgery using the displacement method

Micromanipulation is the precise in vitro handling and study of individual biological cells, where the smallest error can be disastrous. One such example is the extraction of cellular material from multicellular organisms, such as cells from early stage embryos. In this paper, we propose automation methods for the extraction and retrieval of individual cells from a multicellular organism in vitro using the displacement method. Computer-controlled syringe pumps and micromanipulators combined with custom computer vision algorithms are used for automated cell extraction and retrieval. Automation feasibility is demonstrated through automated controlled extraction of one or two blastomeres from cleavage-stage embryos. Preliminary proof of concept blastomere extraction experiments involving mouse embryos obtained success rates ranging from 72% to 88% for the different extraction tasks: displacement, detection, and retrieval. These automated blastomere extraction experiments demonstrate that automated cell extraction is indeed feasible, but the process may still be improved. To the best of these authors' knowledge, this paper is the first to report the automation of single cell extraction from multicellular organisms using the displacement method, and especially for automated blastomere extraction from cleavage-stage embryos. These methods provide a set of tools for moving towards fully automated single cell surgery procedures.

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.

Deep learning enables robust assessment and selection of human blastocysts after in vitro fertilization

Visual morphology assessment is routinely used for evaluating of embryo quality and selecting human blastocysts for transfer after in vitro fertilization (IVF). However, the assessment produces different results between embryologists and as a result, the success rate of IVF remains low. To overcome uncertainties in embryo quality, multiple embryos are often implanted resulting in undesired multiple pregnancies and complications. Unlike in other imaging fields, human embryology and IVF have not yet leveraged artificial intelligence (AI) for unbiased, automated embryo assessment. We postulated that an AI approach trained on thousands of embryos can reliably predict embryo quality without human intervention. We implemented an AI approach based on deep neural networks (DNNs) to select highest quality embryos using a large collection of human embryo time-lapse images (about 50,000 images) from a high-volume fertility center in the United States. We developed a framework (STORK) based on Google’s Inception model. STORK predicts blastocyst quality with an AUC of >0.98 and generalizes well to images from other clinics outside the US and outperforms individual embryologists. Using clinical data for 2182 embryos, we created a decision tree to integrate embryo quality and patient age to identify scenarios associated with pregnancy likelihood. Our analysis shows that the chance of pregnancy based on individual embryos varies from 13.8% (age ≥41 and poor-quality) to 66.3% (age <37 and good-quality) depending on automated blastocyst quality assessment and patient age. In conclusion, our AI-driven approach provides a reproducible way to assess embryo quality and uncovers new, potentially personalized strategies to select embryos.

Anti-Oxidative and Anti-Apoptotic Effects of Apigenin on Number of Viable and Apoptotic Blastomeres, Zona Pellucida Thickness and Hatching Rate of Mouse Embryos

Background

Apigenin is a plant-derived compound belonging to the flavonoids category and bears protective effects on different cells. The aim of this study was to evaluate the effect of apigenin on the number of viable and apoptotic blastomeres, the zona pellucida (ZP) thickness and hatching rate of pre-implantation mouse embryos exposed to H₂O₂ and actinomycin D.

Materials and Methods

In this experimental study, 420 two-cell embryos were randomly divided into six groups: i. Control, ii. Apigenin, iii. H₂O₂ , iv. Apigenin+H₂O₂ , v. Actinomycin D, and vi. Apigenin+Actinomycin D. The percentage of blastocysts and hatched blastocysts was calculated. Blastocyst ZP thickness was also measured. In addition, viable blastomeres quantity was counted by Hoechst and propidium iodide staining and the number of apoptotic blastomeres was counted by TUNEL assay.

Results

The results of viable and apoptotic blastomeres quantity, the ZP thickness, and the percentage of blastocysts and hatched blastocysts were significantly more favorable in the apigenin group, rather than the control group (P<0.05). The results of the apigenin+H₂O₂ group were significantly more favorable than the H₂O₂ group (P<0.05); and the results of apigenin+actinomycin D group were significantly more favorable than actinomycin D group (P<0.05).

Conclusion

The results suggest that apigenin may protect mouse embryos against H₂O₂ and actinomycin D. So that it increases the number of viable blastomeres and decreases the number of apoptotic blastomeres, which may cause expanding the blastocysts, thinning of the ZP thickness and increasing the rate of hatching in mouse embryos.

Effect of quercetin on the number of blastomeres, zona pellucida thickness, and hatching rate of mouse embryos exposed to actinomycin D: An experimental study

BACKGROUND

Quercetin is a flavonoid with the ability to improve the growth of embryos in vitro, and actinomycin D is an inducer of apoptosis in embryonic cells.

OBJECTIVE

The aim was to evaluate the effect of quercetin on the number of viable and apoptotic cells, the zona pellucida (ZP) thickness and the hatching rate of preimplantation embryos exposed to actinomycin D in mice.

MATERIALS AND METHODS

Two-cell embryos were randomly divided into four groups (Control, Quercetin, actinomycin D, and Quercetin + actinomycin D group). Blastocysts percentage, hatched blastocysts, and ZP thickness of blastocysts was measured. The number of blastomeres was counted by Hoechst and propidium iodide staining and the apoptotic cells number was counted by TUNEL assay.

RESULTS

The results showed that the use of quercetin significantly improved the growth of embryos compared to the control group (p=0.037). Moreover, quercetin reduced the destructive effects of actinomycin D on the growth of embryos significantly (p=0.026).

CONCLUSION

quercetin may protect the embryos against actinomycin D so that increases the number of viable cells and decreases the number of apoptotic cells, which can help the expansion of the blastocysts, thinning of the ZP thickness and increasing the hatching rate in mouse embryos.

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.

Predicting pregnancy rate following multiple embryo transfers using algorithms developed through static image analysis

Single-embryo image assessment involves a high degree of inaccuracy because of the imprecise labelling of the transferred embryo images. In this study, we considered the entire transfer cycle to predict the implantation potential of embryos, and propose a novel algorithm based on a combination of local binary pattern texture feature and Adaboost classifiers to predict pregnancy rate. The first step of the proposed method was to extract the features of the embryo images using the local binary pattern operator. After this, multiple embryo images in a transfer cycle were considered as one entity, and the pregnancy rate was predicted using three classifiers: the Real Adaboost, Gentle Adaboost, and Modest Adaboost. Finally, the pregnancy rate was determined via the majority vote rule based on classification results of the three Adaboost classifiers. The proposed algorithm was verified to have a good predictive performance and may assist the embryologist and clinician to select embryos to transfer and in turn improve pregnancy rate.

Detecting Ellipses in Embryo Images Using Arc Detection Method with Particle Swarm for Blastomere-Quality Measurement System

The objective of this paper is to present a novel method, based on a swarm intelligence algorithm, for ellipse detection in digital images of embryo. The process is carried out in several stages. First, edge detection is performed on the image. Then, line segments in the image are detected, and potential elliptical arc segments are extracted from the line segments. Afterward, the detection process is carried out using the Particle Swarm Optimization (PSO) method, which utilize the calculation of the fitness function from the arc segment previously detected. The PSO technique, which is the idea behind the proposed algorithm, is used to find the actual ellipses by combining potential elliptical arcs. The best combination of potential arcs is determined by means a voting technique that utilizes three important points on the arc, namely the starting point, midpoint, and endpoint, so the voting is more efficient than doing the voting for every single pixel in the image. Furthermore, this method is used an embryo image that has following the characteristics: multiple ellipses, a lot of noise, an incomplete ellipse, low image contrast, and overlapping cells. Experiment show that the proposed method detects the ellipses better than do several voting-based ellipse detection methods such as RHT, IRHT, and PSORHT. On the other hand, the experiments show that the proposed method has a higher average hit rate than do other methods. This research is used to increase the success rate of In-Vitro Fertilization (IVF).

Non-invasive prediction of blastocyst implantation, ongoing pregnancy and live birth, by mass spectrometry lipid fingerprinting

Objective

To identify lipid markers of blastocyst implantation and ongoing pregnancy by day three culture medium mass spectrometry (MS) fingerprinting.

Methods

For this study, 33 culture media samples were harvested on day three, from 22 patients undergoing day five embryo transfers. All embryos achieved the blastocyst stage and were split into groups based on their implantation (Negative Implantation, n= 14 and Positive Implantation, n= 19). The positive implantation cycles resulted in successful ongoing pregnancies. The lipid extraction was performed by the Bligh-Dyer protocol and mass spectra were obtained with a direct infusion into a Q-Tof mass spectrometer. The data obtained was analyzed by Principal Component Analysis (PCA) and Partial Least Square Discrimination Analysis (PLS-DA). The statistical analysis was performed using the Metabo-Analyst 2.0.

Results

The variable importance in the projection (VIP) plot of the PLS-DA provided a list of four ions, in the positive mode, with an area under the curve (AUC) of 73.5%; and eight ions, in the negative mode, with and AUC of 72.0%. For both positive and negative modes, possible biomarkers for the negative implantation were identified by the lipidmaps: phosphoethanolamine, dicarboxylic acids, glycerophosphoglycerol, glycerophosphocholine, glicerophosphoinositol, phosphoethanolamine and unsaturated fat acids. The other ions were not identified. These lipids are involved in the GPI anchor biosynthesis and synthesis of lycerophospholipids and phosphate inositol.

Conclusion

MS fingerprinting is useful to identify blastocysts that fail to implant, and therefore this technique could be incorporated into the laboratory routine, adjunct to morphology evaluation to identify embryos that should not be transferred.

Effect of L-carnitine on in vitro developmental rate, the zona pellucida and hatching of blastocysts and their cell numbers in mouse embryos

L-carnitine (LC) is an antioxidant with the ability to promote the growth in vitro embryo.

OBJECTIVE

The goal was to evaluate the effect of LC on some indicators of embryo development and blastocyst quality including zona pellucid (ZP) thickness, the hatching of blastocysts and their cell numbers.

MATERIALS AND METHODS

Mouse embryos were randomly divided into five groups and incubated with different concentrations of LC (I; 0, II; 0.5, III; 1, IV; 2 and V; 4 mg/ml) from 2-cell to hatched blastocyst. The percentage of blastocysts and hatched blastocysts was calculated. Blastocysts ZP thickness was measured and the number of blastocyst cells was counted using Hoechst and propidium iodide (PI) staining.

RESULTS

The results showed concentration of 0.5 mg/ml of LC had an antioxidant effect as in this group, the percentage of blastocysts and hatched blactocysts (p=0.01), the ZP thickness (p=0.00) and the number of blastocyst inner cell mass were significantly more favorable than the control group (p=0.03); and concentration of 4 mg/ml of LC had a toxic effect on embryo development and blastocyst quality (p=0.00).

CONCLUSION

The results suggest that LC may increase the number of blastocyst cells, which probably helps to expand the blastocyst and thinning of the ZP thickness and, therefore, creating a successful hatching for implantation.

A quantitative approach to blastocyst quality evaluation: morphometric analysis and related IVF outcomes

PURPOSE

To quantify blastocyst morphologic parameters with a feasible and standardized tool, investigating their predictive value on implantation outcome.

METHOD

The study retrospectively analyzes 124 blastocysts from 75 patients. Quantitative measurements of blastocyst expansion, inner cell mass and trophoectoderm were taken using digital image analysis software.

RESULT(S)

Blastocysts areas were found to be ranging from 11626.2 up to 35076.4 μm(2). The area of an early blastocyst is A ≤ 18500 μm(2) with a mean diameter d = 140 ± 9 μm, and the area of an expanded blastocyst is A ≥ 24000 with d = 190 ± 9 μm. While blastocyst mean area was not related to implantation rate, more expanded blastocysts displayed a significantly higher implantation rate. Trophoectoderm cell number is a predictor of positive outcome: since a higher of cells (25.6 ± 11.3 vs 16.3 ± 12.8) `forming a tightly knit epithelium is prognostic of implantation potential. Conversely, inner cell mass size is significantly related to implantation only in expanded blastocysts (3122.7 ± 739.0 vs. 2978.1 ± 366.0 μm(2)).

CONCLUSION(S)

Evaluation of blastocyst morphology with a digital image system could be a valuable tool to standardize blastocyst grading based on quantitative parameters. Therefore, digital analysis may be helpful in identifying the best blastocyst to transfer.

Automatic blastomere recognition from a single embryo image

The number of blastomeres of human day 3 embryos is one of the most important criteria for evaluating embryo viability. However, due to the transparency and overlap of blastomeres, it is a challenge to recognize blastomeres automatically using a single embryo image. This study proposes an approach based on least square curve fitting (LSCF) for automatic blastomere recognition from a single image. First, combining edge detection, deletion of multiple connected points, and dilation and erosion, an effective preprocessing method was designed to obtain part of blastomere edges that were singly connected. Next, an automatic recognition method for blastomeres was proposed using least square circle fitting. This algorithm was tested on 381 embryo microscopic images obtained from the eight-cell period, and the results were compared with those provided by experts. Embryos were recognized with a 0 error rate occupancy of 21.59%, and the ratio of embryos in which the false recognition number was less than or equal to 2 was 83.16%. This experiment demonstrated that our method could efficiently and rapidly recognize the number of blastomeres from a single embryo image without the need to reconstruct the three-dimensional model of the blastomeres first; this method is simple and efficient.

A minimally invasive methodology based on morphometric parameters for day 2 embryo quality assessment

The risk of multiple pregnancy to maternal-fetal health can be minimized by reducing the number of embryos transferred. New tools for selecting embryos with the highest implantation potential should be developed. The aim of this study was to evaluate the ability of morphological and morphometric variables to predict implantation by analysing images of embryos. This was a retrospective study of 135 embryo photographs from 112 IVF-ICSI cycles carried out between January and March 2011. The embryos were photographed immediately before transfer using Cronus 3 software. Their images were analysed using the public program ImageJ. Significant effects (P < 0.05), and higher discriminant power to predict implantation were observed for the morphometric embryo variables compared with morphological ones. The features for successfully implanted embryos were as follows: four cells on day 2 of development; all blastomeres with circular shape (roundness factor greater than 0.9), an average zona pellucida thickness of 13 µm and an average of 17695.1 µm² for the embryo area. Embryo size, which is described by its area and the average roundness factor for each cell, provides two objective variables to consider when predicting implantation. This approach should be further investigated for its potential ability to improve embryo scoring.

A unified graphical models framework for automated mitosis detection in human embryos

Time lapse microscopy has emerged as an important modality for studying human embryo development, as mitosis events can provide insight into embryo health and fate. Mitosis detection can happen through tracking of embryonic cells (tracking based), or from low level image features and classifiers (tracking free). Tracking based approaches are challenged by high dimensional search space, weak features, outliers, missing data, multiple deformable targets, and weak motion model. Tracking free approaches are data driven and complement tracking based approaches. We pose mitosis detection as augmented simultaneous segmentation and classification in a conditional random field (CRF) framework that combines both approaches. It uses a rich set of discriminative features and their spatiotemporal context. It performs a dual pass approximate inference that addresses the high dimensionality of tracking and combines results from both components. For 312 clinical sequences we measured division events to within 30 min and observed an improvement of 25.6% and a 32.9% improvement over purely tracking based and tracking free approach respectively, and close to an order of magnitude over a traditional particle filter. While our work was motivated by human embryo development, it can be extended to other detection problems in image sequences of evolving cell populations.

The importance of the cleavage stage morphology evaluation for blastocyst transfer in patients with good prognosis

PURPOSE

To evaluate: (i) the influence of morphology at cleavage stage on blastocyst formation and implantation, and (ii) whether the transfer of low-quality embryos on day-three would be a better approach than the transfer at blastocyst stage.

METHODS

This study included 8,444 embryos obtained from 1,125 patients undergoing ICSI cycles between January/2011 and September/2013. The influence of the quality of the embryo on days-two and -three on blastocyst formation and implantation success was evaluated. Moreover, the implantation potential of low-quality embryos, at cleavage stage, transferred on day-three was compared with the implantation potential of low-quality embryos, at cleavage stage, transferred on day-five.

RESULTS

Low-quality embryos on day-two had an approximate 20 % decreased chance of achieving the blastocyst stage, and blastocysts derived from low-quality embryos on day-two had a nearly 40 % decrease in the implantation chance. Low-quality embryos on day-three had a 30 % decreased chance of achieving the blastocyst stage, and blastocysts derived from low-quality embryos on day-three had an almost 40 % decreased implantation chance. The implantation rate didn't differ when low-quality embryos on the cleavage stage were transferred on day-three or left in culture and transferred on day-five.

CONCLUSIONS

The transfer of low-quality embryos on day-three is a better approach than transfer at the blastocyst stage. In addition, the embryo morphology evaluation at the cleavage stage is still needed for the selection of the embryo with the best implantation potential in extended embryo culture programmes.

Improving embryo selection using a computer-automated time-lapse image analysis test plus day 3 morphology: results from a prospective multicenter trial

OBJECTIVE

To assess the first computer-automated platform for time-lapse image analysis and blastocyst prediction and to determine how the screening information may assist embryologists in day 3 (D3) embryo selection.

DESIGN

Prospective, multicenter, cohort study.

SETTING

Five IVF clinics in the United States.

PATIENT(S)

One hundred sixty women ≥ 18 years of age undergoing fresh IVF treatment with basal antral follicle count ≥ 8, basal FSH <10 IU/mL, and ≥ 8 normally fertilized oocytes.

INTERVENTION(S)

A noninvasive test combining time-lapse image analysis with the cell-tracking software, Eeva (Early Embryo Viability Assessment), was used to measure early embryo development and generate usable blastocyst predictions by D3.

MAIN OUTCOME MEASURE(S)

Improvement in the ability of experienced embryologists to select which embryos are likely to develop to usable blastocysts using D3 morphology alone, compared with morphology plus Eeva.

RESULT(S)

Experienced embryologists using Eeva in combination with D3 morphology significantly improved their ability to identify embryos that would reach the usable blastocyst stage (specificity for each of three embryologists using morphology vs. morphology plus Eeva: 59.7% vs. 86.3%, 41.9% vs. 84.0%, 79.5% vs. 86.6%). Adjunctive use of morphology plus Eeva improved embryo selection by enabling embryologists to better discriminate which embryos would be unlikely to develop to blastocyst and was particularly beneficial for improving selection among good-morphology embryos. Adjunctive use of morphology plus Eeva also reduced interindividual variability in embryo selection.

CONCLUSION(S)

Previous studies have shown improved implantation rates for blastocyst transfer compared with cleavage-stage transfer. Addition of Eeva to the current embryo grading process may improve the success rates of cleavage-stage ETs.

A method for semi-automatic grading of human blastocyst microscope images

BACKGROUND

The precise assessment of embryo viability is an extremely important factor for the optimization of IVF treatments. In order to assess embryo viability, several embryo scoring systems have been developed. However, they rely mostly on a subjective visual analysis of embryo morphological features and thus are subject to inter- and intra-observer variation. In this paper, we propose a method for image segmentation (the dividing of an image into its meaningful constituent regions) and classification of human blastocyst images with the aim of automating embryo grading.

METHODS

The delineation of the boundaries (segmentation) of the zona pellucida, trophectoderm (TE) and inner cell mass (ICM) were performed using advanced image analysis techniques (level set, phase congruency and fitting of ellipse methods). The fractal dimension and mean thickness of TE and ICM image texture descriptors (texture spectrum and grey-level run lengths) were calculated to characterize the main morphological features of the blastocyst with the aim of automatic grading using Support Vector Machine classifiers.

RESULTS

The fractal dimension calculated from the delineated TE boundary provided a good indication of cell number (presented a 0.81 Pearson correlation coefficient with the number of cells), a feature closely associated with blastocyst quality. The classifiers showed different accuracy levels for each grade. They presented accuracy ranges from 0.67 to 0.92 for the embryo development classification, 0.67-0.82 for the ICM classification and 0.53-0.92 for the TE classification. The value 0.92 was the highest accuracy achieved in the tests with 73 blastocysts.

CONCLUSIONS

Semi-automatic grading of human blastocysts by a computer is feasible and may offer a more precise comparison of embryos, reducing subjectivity and allowing embryos with apparently identical morphological scores to be distinguished.