Analysis and quantification of female and male contributions to the first stages of embryonic kinetics: study from a time-lapse system.

PURPOSE: The few studies that examined the effect of male and/or female features on early embryo development, notably using the time-lapse system (TL), reported conflicting results. This can be explained by the small number of studies using an adapted model. METHODS: We used two original designs to study the female and male effects on embryo development: (1) based on embryos from donor oocytes (TL-DO), and (2) from donor sperm (TL-DS). Firstly, we analyzed the female and male similarities using an ad hoc intraclass correlation coefficient (ICC), then we completed the analysis with a multivariable model to assess the association between both male and female factors, and early embryo kinetics. A total of 572 mature oocytes (TL-DO: 293; TL-DS: 279), fertilized by intracytoplasmic sperm injection (ICSI) and incubated in a TL (Embryoscope®) were included from March 2013 to April 2019; 429 fertilized oocytes (TL-DO: 212; TL-DS: 217) were assessed. The timings of the first 48 h have been analyzed. RESULTS: The similarities in the timings thought to be related to the female component were significant: (ICC in both DO-DS designs respectively: tPB2: 9-18%; tPNa: 16-21%; tPNf: 40-26%; t2: 38-24%; t3: 15-20%; t4: 21-32%). Comparatively, those related to male were lower. Surprisingly after multivariable analyses, no intrinsic female factors were clearly identified. However, in TL-DO design, oligozoospermia was associated with a tendency to longer timings, notably for tPB2 (p = 0.026). CONCLUSION: This study quantifies the role of the oocyte in the first embryo cleavages, but without identified specific female factors. However, it also highlights that sperm may have an early embryonic effect.

Adaptive adversarial neural networks for the analysis of lossy and domain-shifted datasets of medical images.

In machine learning for image-based medical diagnostics, supervised convolutional neural networks are typically trained with large and expertly annotated datasets obtained using high-resolution imaging systems. Moreover, the network's performance can degrade substantially when applied to a dataset with a different distribution. Here, we show that adversarial learning can be used to develop high-performing networks trained on unannotated medical images of varying image quality. Specifically, we used low-quality images acquired using inexpensive portable optical systems to train networks for the evaluation of human embryos, the quantification of human sperm morphology and the diagnosis of malarial infections in the blood, and show that the networks performed well across different data distributions. We also show that adversarial learning can be used with unlabelled data from unseen domain-shifted datasets to adapt pretrained supervised networks to new distributions, even when data from the original distribution are not available. Adaptive adversarial networks may expand the use of validated neural-network models for the evaluation of data collected from multiple imaging systems of varying quality without compromising the knowledge stored in the network.

Comparison of the pregnancy and obstetric outcomes between single cleavage-stage embryo transfer and single blastocyst transfer by time-lapse selection of embryos.

To compare the pregnancy and obstetric outcomes following single cleavage-stage embryo transfer (SCT) and single blastocyst transfer (SBT) using time-lapse imaging (TLI), a total of 2066 normally fertilized and cleaved embryos from 233 patients were divided into Day 3 SCT group (n = 171) and Day 5 SBT group (n = 62) according to patient's willingness. Embryo selection criteria were based on embryo cleavage patterns, timing parameters, and blastocyst quality. The pregnancy and obstetric outcomes of each group were evaluated. There were no statistically significant differences with regard to pregnancy outcomes including the implantation rate, early abortion rate, ongoing pregnancy rate and live birth rate, and obstetric outcomes including preterm birth rate, gestational week, birth height, birth weight and fetal malformation rate between SCT group and SBT group. SBT group had significantly higher monozygotic twinning (MZT) rates than SCT group (6.98% vs. 0, p < .05). Although not statistically significant, there was a trend of higher proportion of male-to-female sex ratio at birth in SBT group than SCT group (1.38 vs. 1.05). Based on the combination of cleavage patterns and timing parameters, SCT may be an alternative to SBT because it can provide similar pregnancy and obstetric outcomes and meanwhile lower monozygotic twinning rates.

Computational geometry analysis of dendritic spines by structured illumination microscopy.

Dendritic spines are the postsynaptic sites that receive most of the excitatory synaptic inputs, and thus provide the structural basis for synaptic function. Here, we describe an accurate method for measurement and analysis of spine morphology based on structured illumination microscopy (SIM) and computational geometry in cultured neurons. Surface mesh data converted from SIM images were comparable to data reconstructed from electron microscopic images. Dimensional reduction and machine learning applied to large data sets enabled identification of spine phenotypes caused by genetic mutations in key signal transduction molecules. This method, combined with time-lapse live imaging and glutamate uncaging, could detect plasticity-related changes in spine head curvature. The results suggested that the concave surfaces of spines are important for the long-term structural stabilization of spines by synaptic adhesion molecules.

Temperature imaging using a cationic linear fluorescent polymeric thermometer and fluorescence lifetime imaging microscopy.

Temperature is one of the most important of the physiological parameters that determine the biological status of living organisms. However, intracellular temperature was not imaged at the single-cell level until recently because of the lack of a molecular thermometer that can be applied to living cells. We have recently developed a method for imaging intracellular temperature using a cationic linear fluorescent polymeric thermometer (FPT) and fluorescence lifetime imaging microscopy (FLIM). The cationic linear FPT exhibits cell permeability in various mammalian cell lines and yeast cells, entering live cells within 10 min of incubation. Intracellular thermometry using the cationic linear FPT and FLIM can be used to image temperature with high temperature resolution (0.3-1.29 °C within a temperature range of 25-35 °C). The diffuse intracellular localization of the cationic linear FPT allows a high spatial resolution (i.e., the light microscope's diffraction limit, 200 nm), enabling the detection of temperature distributions at the subcellular level. This protocol, including the construction of a calibration curve and intracellular temperature imaging, requires ~14 h. Experience in handling cultured mammalian cells and use of a confocal laser-scanning microscope (CLSM) is required.

Dynamic nanoscale morphology of the ER surveyed by STED microscopy.

The endoplasmic reticulum (ER) is composed of interconnected membrane sheets and tubules. Superresolution microscopy recently revealed densely packed, rapidly moving ER tubules mistaken for sheets by conventional light microscopy, highlighting the importance of revisiting classical views of ER structure with high spatiotemporal resolution in living cells. In this study, we use live-cell stimulated emission depletion (STED) microscopy to survey the architecture of the ER at 50-nm resolution. We determine the nanoscale dimensions of ER tubules and sheets for the first time in living cells. We demonstrate that ER sheets contain highly dynamic, subdiffraction-sized holes, which we call nanoholes, that coexist with uniform sheet regions. Reticulon family members localize to curved edges of holes within sheets and are required for their formation. The luminal tether Climp63 and microtubule cytoskeleton modulate their nanoscale dynamics and organization. Thus, by providing the first quantitative analysis of ER membrane structure and dynamics at the nanoscale, our work reveals that the ER in living cells is not limited to uniform sheets and tubules; instead, we suggest the ER contains a continuum of membrane structures that includes dynamic nanoholes in sheets as well as clustered tubules.

Obstetric and perinatal outcomes of pregnancies conceived with embryos cultured in a time-lapse monitoring system.

OBJECTIVE: To compare obstetric and perinatal outcomes of singleton pregnancies resulting from embryos incubated in a time-lapse system (TLS) with those of embryos grown in standard IVF incubators (SI). DESIGN: Retrospective description of a cohort of patients who conceived during a randomized, controlled trial. SETTING: Private university-affiliated IVF center. PATIENT(S): Of 856 randomized patients, 378 gave birth to a live-born infant: 216 of the deliveries originated from embryos incubated in TLS, and 162 deliveries were from embryos cultured in SI. INTERVENTION(S): Embryo incubation and selection in TLS. MAIN OUTCOME MEASURE(S): Delivery and neonatal outcomes. RESULT(S): No significant differences were observed in the baseline characteristics of the study population. The delivery rate was 49.3% (TLS) vs. 40.0% (SI), and multiple deliveries were higher in the TLS group: 31.0% (67 of 216) vs. 24.7% (40 of 162) in the SI group. When singleton pregnancies were analyzed no differences were found between the two groups in the rate of obstetric problems with respect to weeks at delivery: 38.8 (95% confidence interval [CI] 38.4-39.1) (TLS) vs. 39.5 (95% CI 38.0-39.9) (SI); preterm births (<37 weeks): 10.7% (TLS) vs. 12.3% (SI); and very preterm births (<34 weeks): 2.9% (TLS) vs. 3.3% (SI). No statistical differences were found in neonatal outcomes such as birth weight: 3,163 g (95% CI 3,035-3,292 g) (TLS) vs. 3,074 (95% CI 2,913-3,236) (SI); low birth weight (<2,500 g): 12.8% (TLS) vs. 12.3% (SI); very low birth weight (<1,500 g): 2.0% (TLS) vs. 2.4% (SI); or height: 50.3 cm (95% CI 49.6-50.9 cm) (TLS) vs. 49.7 (95% CI 48.9-50.4 cm) (SI). No major malformations or perinatal mortality were found in either of the two groups. CONCLUSION(S): No detrimental effects were observed in obstetric and perinatal outcomes when a time-lapse incubator was used rather than a more widely used conventional incubator. As far as we know this is the first report from a randomized study of the neonatal outcomes of time-lapse monitoring. Our results suggest that this technology is an effective and safe alternative for embryo incubation, though trials of larger numbers of patients are required to further confirm our conclusions. CLINICAL TRIAL REGISTRATION NUMBER: NCT01549262.

A Novel Computer-Assisted Approach to evaluate Multicellular Tumor Spheroid Invasion Assay.

Multicellular tumor spheroids (MCTSs) embedded in a matrix are re-emerging as a powerful alternative to monolayer-based cultures. The primary information gained from a three-dimensional model is the invasiveness of treatment-exposed MCTSs through the acquisition of light microscopy images. The amount and complexity of the acquired data and the bias arisen by their manual analysis are disadvantages calling for an automated, high-throughput analysis. We present a universal algorithm we developed with the scope of being robust enough to handle images of various qualities and various invasion profiles. The novelty and strength of our algorithm lie in: the introduction of a multi-step segmentation flow, where each step is optimized for each specific MCTS area (core, halo, and periphery); the quantification through the density of the two-dimensional representation of a three-dimensional object. This latter offers a fine-granular differentiation of invasive profiles, facilitating a quantification independent of cell lines and experimental setups. Progression of density from the core towards the edges influences the resulting density map thus providing a measure no longer dependent on the sole area size of MCTS, but also on its invasiveness. In sum, we propose a new method in which the concept of quantification of MCTS invasion is completely re-thought.

MorphoGraphX: A platform for quantifying morphogenesis in 4D.

Morphogenesis emerges from complex multiscale interactions between genetic and mechanical processes. To understand these processes, the evolution of cell shape, proliferation and gene expression must be quantified. This quantification is usually performed either in full 3D, which is computationally expensive and technically challenging, or on 2D planar projections, which introduces geometrical artifacts on highly curved organs. Here we present MorphoGraphX ( www.MorphoGraphX.org), a software that bridges this gap by working directly with curved surface images extracted from 3D data. In addition to traditional 3D image analysis, we have developed algorithms to operate on curved surfaces, such as cell segmentation, lineage tracking and fluorescence signal quantification. The software's modular design makes it easy to include existing libraries, or to implement new algorithms. Cell geometries extracted with MorphoGraphX can be exported and used as templates for simulation models, providing a powerful platform to investigate the interactions between shape, genes and growth.

Morphokinetic parameters of ICSI tripronucleated embryos observed using time lapse.

Time-lapse analysis of tripronucleated zygotes obtained in ICSI cycles showed that 75.4% cleaved into embryos. These embryos subsequently demonstrated slower developmental kinetics than normally fertilized embryos.

The use of morphokinetic parameters to select all embryos with full capacity to implant.

PURPOSE: Embryo kinetics analysis is an emerging tool for selecting embryo(s) for transfer. The aim of the present study was to determine morphokinetic parameters easily usable in the laboratory and predictive of embryo development and, most importantly, of embryo competence in producing a clinical pregnancy after day 5 transfer. METHODS: A retrospective time-lapse monitoring analysis of morphokinetic parameters for 72 fully implanted embryos (group A) were compared to 106 non-implanted embryos (group B), and to 66 embryos with arrested development from the same pool of group A. All the embryos were from 78 patients undergoing ICSI treatment and day 5 embryo transfers. RESULTS: A day 3 embryo will develop into a viable blastocyst if the following ranges of morphokinetic parameters are met: t1 (between 18.4 h and 30.9 h post-ICSI), t2 (21.4-34.8 h), t4 (33.1-57.2 h), t7 (46.1-82.5 h), t8 (46.4-97.8 h), tC-tF (7.7-22.9 h) and s3 (0.7-30.8 h). On day 5 embryos with the highest probability to implant are those with a cc3 between 9.7 h and 21 h. CONCLUSIONS: Morphokinetic parameters are helpful to make appropriate decisions for the disposition of each embryo. It is recommended that each laboratory should determine its own ranges of in vitro development (IVD-MKP) and implantation-associated (IMP-MKP) morphokinetic parameters.

Embryo incubation and selection in a time-lapse monitoring system improves pregnancy outcome compared with a standard incubator: a retrospective cohort study.

OBJECTIVE: To quantify the effect on reproductive outcome of culturing and selecting embryos using a novel time-lapse monitoring system (TMS). DESIGN: Retrospective observational cohort study. SETTING: University-affiliated private center. PATIENT(S): Donation and autologous intracytoplasmic sperm injection (ICSI) cycles from ten IVF clinics using similar procedures, cultured in TMS (n = 1,390) or in a standard incubator (SI; n = 5,915). INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Clinical pregnancy rate confirmed by ultrasound in week 7. RESULT(S): A logistic regression analysis, which included all significant confounding factors, was used to evaluate the effect of culturing and selecting embryos with the use of TMS. Comparing clinical pregnancy rates per oocyte retrieval with TMS and SI treatments gave a crude effect of odds ratio [OR] 1.190 (95% confidence interval [CI] 1.058-1.337). Oocyte source, maternal age, day of transfer, and number of retrieved oocytes were identified as significant confounding factors. After accounting for confounding factors, the effect of TMS culture was OR 1.201 (95% CI 1.059-1.363). Limiting analysis to treatments with embryo transfer and including number of transferred embryos as a confounding factor likewise gave a significant effect of TMS with OR 1.157 (95% CI 1.018-1.315). CONCLUSION(S): Analysis of retrospective data indicated that culturing and selecting embryos by TMS significantly improved the relative probability of clinical pregnancy (+20.1% per oocyte retrieval, +15.7% per embryo transfer). The elevated clinical pregnancy rate was attributed to a combination of stable culture conditions and the use of morphokinetic parameters for embryo selection.