Causal inference indicates that poor responders have similar outcomes with the antagonist protocol compared with flare.

OBJECTIVE: To use causal inference to investigate whether the flare or antagonist protocol is better for poor responders going through controlled ovarian stimulation. DESIGN: A retrospective study. SETTING: Retrieval cycles from the Society for Assisted Reproductive Technology Clinic Outcomes Reporting System. PATIENTS: Patients in the United States underwent autologous in vitro fertilization cycles from 2014 to 2019 using either the flare or antagonist protocol. INTERVENTION: Not applicable. MAIN OUTCOME MEASURE: Primary outcomes included oocytes retrieved, fertilized oocytes (2PNs), blastocysts, the cumulative live birth rate (CLBR), and cycle cancelation rate. RESULTS: After propensity score matching, patients with a predicted poor response (antimüllerian hormone, <0.5) on their first in vitro fertilization cycle had similar outcomes on the antagonist protocol (CLBR of 14.2%, 95% confidence intervals [CIs]: 13.6%, 14.8%) compared with flare (CLBR of 13.6%, 95% CIs: 12.4%, 14.8%). We evaluated patients undergoing a second cycle after having a poor response (<4 oocytes retrieved) on their first cycle. Patients in the antagonist-to-antagonist group had a similar change in outcomes between the first and second cycles (average CLBR improvement of 13.9%, 95% CIs: 12.1%, 15.6%) compared with the antagonist-to-flare group (average CLBR improvement of 14.4%, 95% CIs: 10.9%, 18.3%). In addition, patients in the flare-to-antagonist group had a similar change in outcomes between the first and second cycles (average CLBR improvement of 10.4%, 95% CIs: 6.6%, 14.5%) compared with the flare-to-flare group (average CLBR improvement of 9.0%, 95% CIs: 5.1%, 13.4%). CONCLUSION: Poor responders have similar outcomes on an antagonist protocol compared with a flare protocol for both the first and second cycles.

A higher number of oocytes retrieved is associated with an increase in fertilized oocytes, blastocysts, and cumulative live birth rates.

OBJECTIVE: To investigate the association between the number of oocytes retrieved and the numbers of fertilized oocytes and blastocysts and cumulative and primary transfer live birth rates (LBRs). DESIGN: Retrospective study. SETTING: Retrieval cycles and linked embryo transfers from the Society for Assisted Reproductive Technology Clinic Outcome Reporting System. PATIENT(S): Patients in the United States undergoing autologous in vitro fertilization cycles from 2014 to 2019 (n = 402,411 cycles). INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Normally fertilized oocytes, blastocysts, and cumulative and primary transfer LBRs. RESULT(S): There was a strong positive linear correlation between oocytes and fertilized oocytes and between oocytes and blastocysts. The cumulative LBR increased rapidly with the number of oocytes retrieved to approximately 16-20 oocytes, at which point it continued to increase but with diminishing returns. The increasing trend of the cumulative LBR was observed when stratifying patients by age and antimüllerian hormone and after controlling for confounding variables using multivariate logistic regression. The primary transfer LBR also increased with the number of oocytes to approximately 16-20 oocytes, at which point it plateaued but did not decline. CONCLUSION(S): A higher number of oocytes retrieved improves the cumulative LBR without impairing the primary transfer LBR. This suggests that ovarian stimulation strategies should aim to safely maximize the number of oocytes retrieved.

An interpretable machine learning model for individualized gonadotrophin starting dose selection during ovarian stimulation.

RESEARCH QUESTION: Can we develop an interpretable machine learning model that optimizes starting gonadotrophin dose selection in terms of mature oocytes (metaphase II [MII]), fertilized oocytes (2 pronuclear [2PN]) and usable blastocysts? DESIGN: This was a retrospective study of patients undergoing autologous IVF cycles from 2014 to 2020 (n = 18,591) in three assisted reproductive technology centres in the USA. For each patient cycle, an individual dose-response curve was generated from the 100 most similar patients identified using a K-nearest neighbours model. Patients were labelled as dose-responsive if their dose-response curve showed a region that maximized MII oocytes, and flat-responsive otherwise. RESULTS: Analysis of the dose-response curves showed that 30% of cycles were dose-responsive and 64% were flat-responsive. After propensity score matching, patients in the dose-responsive group who received an optimal starting dose of FSH had on average 1.5 more MII oocytes, 1.2 more 2PN embryos and 0.6 more usable blastocysts using 10 IU less of starting FSH and 195 IU less of total FSH compared with patients given non-optimal doses. In the flat-responsive group, patients who received a low starting dose of FSH had on average 0.3 more MII oocytes, 0.3 more 2PN embryos and 0.2 more usable blastocysts using 149 IU less of starting FSH and 1375 IU less of total FSH compared with patients with a high starting dose. CONCLUSIONS: This study demonstrates retrospectively that using a machine learning model for selecting starting FSH can achieve optimal laboratory outcomes while reducing the amount of starting and total FSH used.

Phenotypic screening with deep learning identifies HDAC6 inhibitors as cardioprotective in a BAG3 mouse model of dilated cardiomyopathy.

Dilated cardiomyopathy (DCM) is characterized by reduced cardiac output, as well as thinning and enlargement of left ventricular chambers. These characteristics eventually lead to heart failure. Current standards of care do not target the underlying molecular mechanisms associated with genetic forms of heart failure, driving a need to develop novel therapeutics for DCM. To identify candidate therapeutics, we developed an in vitro DCM model using induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs) deficient in B-cell lymphoma 2 (BCL2)-associated athanogene 3 (BAG3). With these BAG3-deficient iPSC-CMs, we identified cardioprotective drugs using a phenotypic screen and deep learning. From a library of 5500 bioactive compounds and siRNA validation, we found that inhibiting histone deacetylase 6 (HDAC6) was cardioprotective at the sarcomere level. We translated this finding to a BAG3 cardiomyocyte-knockout (BAG3cKO) mouse model of DCM, showing that inhibiting HDAC6 with two isoform-selective inhibitors (tubastatin A and a novel inhibitor TYA-018) protected heart function. In BAG3cKO and BAG3E455K mice, HDAC6 inhibitors improved left ventricular ejection fraction and reduced left ventricular diameter at diastole and systole. In BAG3cKO mice, TYA-018 protected against sarcomere damage and reduced Nppb expression. Based on integrated transcriptomics and proteomics and mitochondrial function analysis, TYA-018 also enhanced energetics in these mice by increasing expression of targets associated with fatty acid metabolism, protein metabolism, and oxidative phosphorylation. Our results demonstrate the power of combining iPSC-CMs with phenotypic screening and deep learning to accelerate drug discovery, and they support developing novel therapies that address underlying mechanisms associated with heart disease.

An interpretable machine learning model for predicting the optimal day of trigger during ovarian stimulation.

OBJECTIVE: To develop an interpretable machine learning model for optimizing the day of trigger in terms of mature oocytes (MII), fertilized oocytes (2PNs), and usable blastocysts. DESIGN: Retrospective study. SETTING: A group of three assisted reproductive technology centers in the United States. PATIENT(S): Patients undergoing autologous in vitro fertilization cycles from 2014 to 2020 (n = 30,278). INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Average number of MII oocytes, 2PNs, and usable blastocysts. RESULT(S): A set of interpretable machine learning models were developed using linear regression with follicle counts and estradiol levels. When using the model to make day-by-day predictions of trigger or continuing stimulation, possible early and late triggers were identified in 48.7% and 13.8% of cycles, respectively. After propensity score matching, patients with early triggers had on average 2.3 fewer MII oocytes, 1.8 fewer 2PNs, and 1.0 fewer usable blastocysts compared with matched patients with on-time triggers, and patients with late triggers had on average 2.7 fewer MII oocytes, 2.0 fewer 2PNs, and 0.7 fewer usable blastocysts compared with matched patients with on-time triggers. CONCLUSION(S): This study demonstrates that it is possible to develop an interpretable machine learning model for optimizing the day of trigger. Using our model has the potential to improve outcomes for many in vitro fertilization patients.

Characterization of an artificial intelligence model for ranking static images of blastocyst stage embryos.

OBJECTIVE: To perform a series of analyses characterizing an artificial intelligence (AI) model for ranking blastocyst-stage embryos. The primary objective was to evaluate the benefit of the model for predicting clinical pregnancy, whereas the secondary objective was to identify limitations that may impact clinical use. DESIGN: Retrospective study. SETTING: Consortium of 11 assisted reproductive technology centers in the United States. PATIENT(S): Static images of 5,923 transferred blastocysts and 2,614 nontransferred aneuploid blastocysts. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Prediction of clinical pregnancy (fetal heartbeat). RESULT(S): The area under the curve of the AI model ranged from 0.6 to 0.7 and outperformed manual morphology grading overall and on a per-site basis. A bootstrapped study predicted improved pregnancy rates between +5% and +12% per site using AI compared with manual grading using an inverted microscope. One site that used a low-magnification stereo zoom microscope did not show predicted improvement with the AI. Visualization techniques and attribution algorithms revealed that the features learned by the AI model largely overlap with the features of manual grading systems. Two sources of bias relating to the type of microscope and presence of embryo holding micropipettes were identified and mitigated. The analysis of AI scores in relation to pregnancy rates showed that score differences of ≥0.1 (10%) correspond with improved pregnancy rates, whereas score differences of <0.1 may not be clinically meaningful. CONCLUSION(S): This study demonstrates the potential of AI for ranking blastocyst stage embryos and highlights potential limitations related to image quality, bias, and granularity of scores.

Deep learning detects cardiotoxicity in a high-content screen with induced pluripotent stem cell-derived cardiomyocytes.

Drug-induced cardiotoxicity and hepatotoxicity are major causes of drug attrition. To decrease late-stage drug attrition, pharmaceutical and biotechnology industries need to establish biologically relevant models that use phenotypic screening to detect drug-induced toxicity in vitro. In this study, we sought to rapidly detect patterns of cardiotoxicity using high-content image analysis with deep learning and induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs). We screened a library of 1280 bioactive compounds and identified those with potential cardiotoxic liabilities in iPSC-CMs using a single-parameter score based on deep learning. Compounds demonstrating cardiotoxicity in iPSC-CMs included DNA intercalators, ion channel blockers, epidermal growth factor receptor, cyclin-dependent kinase, and multi-kinase inhibitors. We also screened a diverse library of molecules with unknown targets and identified chemical frameworks that show cardiotoxic signal in iPSC-CMs. By using this screening approach during target discovery and lead optimization, we can de-risk early-stage drug discovery. We show that the broad applicability of combining deep learning with iPSC technology is an effective way to interrogate cellular phenotypes and identify drugs that may protect against diseased phenotypes and deleterious mutations.

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.

Quantifying drug-induced structural toxicity in hepatocytes and cardiomyocytes derived from hiPSCs using a deep learning method.

Cardiac and hepatic toxicity result from induced disruption of the functioning of cardiomyocytes and hepatocytes, respectively, which is tightly related to the organization of their subcellular structures. Cellular structure can be analyzed from microscopy imaging data. However, subtle or complex structural changes that are not easily perceived may be missed by conventional image-analysis techniques. Here we report the evaluation of PhenoTox, an image-based deep-learning method of quantifying drug-induced structural changes using human hepatocytes and cardiomyocytes derived from human induced pluripotent stem cells. We assessed the ability of the deep learning method to detect variations in the organization of cellular structures from images of fixed or live cells. We also evaluated the power and sensitivity of the method for detecting toxic effects of drugs by conducting a set of experiments using known toxicants and other methods of screening for cytotoxic effects. Moreover, we used PhenoTox to characterize the effects of tamoxifen and doxorubicin-which cause liver toxicity-on hepatocytes. PhenoTox revealed differences related to loss of cytochrome P450 3A4 activity, for which it showed greater sensitivity than a caspase 3/7 assay. Finally, PhenoTox detected structural toxicity in cardiomyocytes, which was correlated with contractility defects induced by doxorubicin, erlotinib, and sorafenib. Taken together, the results demonstrated that PhenoTox can capture the subtle morphological changes that are early signs of toxicity in both hepatocytes and cardiomyocytes.

A non-invasive platform for functional characterization of stem-cell-derived cardiomyocytes with applications in cardiotoxicity testing.

We present a non-invasive method to characterize the function of pluripotent stem-cell-derived cardiomyocytes based on video microscopy and image analysis. The platform, called Pulse, generates automated measurements of beating frequency, beat duration, amplitude, and beat-to-beat variation based on motion analysis of phase-contrast images captured at a fast frame rate. Using Pulse, we demonstrate recapitulation of drug effects in stem-cell-derived cardiomyocytes without the use of exogenous labels and show that our platform can be used for high-throughput cardiotoxicity drug screening and studying physiologically relevant phenotypes.

Automated, non-invasive characterization of stem cell-derived cardiomyocytes from phase-contrast microscopy.

Stem cell-derived cardiomyocytes hold tremendous potential for drug development and safety testing related to cardiovascular health. The characterization of cardiomyocytes is most commonly performed using electrophysiological systems, which are expensive, laborious to use, and may induce undesirable cellular response. Here, we present a new method for non-invasive characterization of cardiomyocytes using video microscopy and image analysis. We describe an automated pipeline that consists of segmentation of beating regions, robust beating signal calculation, signal quantification and modeling, and hierarchical clustering. Unlike previous imaging-based methods, our approach enables clinical applications by capturing beating patterns and arrhythmias across healthy and diseased cells with varied densities. We demonstrate the strengths of our algorithm by characterizing the effects of two commercial drugs known to modulate beating frequency and irregularity. Our results provide, to our knowledge, the first clinically-relevant demonstration of a fully-automated and non-invasive imaging-based beating assay for characterization of stem cell-derived cardiomyocytes.

A system for automated, noninvasive, morphology-based evaluation of induced pluripotent stem cell cultures.

Due to the rapid adoption and use of human induced pluripotent stem cells (iPSCs) in recent years, there is a need for new technologies that standardize the evaluation of iPSCs to allow the objective comparison of results across different experiments and groups. In this article, we present a noninvasive, fully automated, and analytical system for morphology-based evaluation of iPSC cultures that consists of time-lapse microscopy and novel image analysis software. The presented system acquires low-light phase-contrast images of iPSC growth collected during a period of several days in culture, measures geometrical- and texture-based features of iPSC colonies throughout time, and derives a set of six biologically relevant features to automatically rank the quality of the cell culture. In a study of 94 iPSC cultures, we demonstrated the accuracy of the system by comparing the automated ranking with an independent expert evaluation based on visual review of the time-lapse movies. To our knowledge, this is the first demonstration of a fully automated and objective assessment of iPSC culture quality using noninvasive methods.

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.

Dynamic blastomere behaviour reflects human embryo ploidy by the four-cell stage.

Previous studies have demonstrated that aneuploidy in human embryos is surprisingly frequent with 50-80% of cleavage-stage human embryos carrying an abnormal chromosome number. Here we combine non-invasive time-lapse imaging with karyotypic reconstruction of all blastomeres in four-cell human embryos to address the hypothesis that blastomere behaviour may reflect ploidy during the first two cleavage divisions. We demonstrate that precise cell cycle parameter timing is observed in all euploid embryos to the four-cell stage, whereas only 30% of aneuploid embryos exhibit parameter values within normal timing windows. Further, we observe that the generation of human embryonic aneuploidy is complex with contribution from chromosome-containing fragments/micronuclei that frequently emerge and may persist or become reabsorbed during interphase. These findings suggest that cell cycle and fragmentation parameters of individual blastomeres are diagnostic of ploidy, amenable to automated tracking algorithms, and likely of clinical relevance in reducing transfer of embryos prone to miscarriage.

In vivo near-infrared dual-axis confocal microendoscopy in the human lower gastrointestinal tract.

Near-infrared confocal microendoscopy is a promising technique for deep in vivo imaging of tissues and can generate high-resolution cross-sectional images at the micron-scale. We demonstrate the use of a dual-axis confocal (DAC) near-infrared fluorescence microendoscope with a 5.5-mm outer diameter for obtaining clinical images of human colorectal mucosa. High-speed two-dimensional en face scanning was achieved through a microelectromechanical systems (MEMS) scanner while a micromotor was used for adjusting the axial focus. In vivo images of human patients are collected at 5 frames/sec with a field of view of 362×212 µm(2) and a maximum imaging depth of 140 µm. During routine endoscopy, indocyanine green (ICG) was topically applied a nonspecific optical contrasting agent to regions of the human colon. The DAC microendoscope was then used to obtain microanatomic images of the mucosa by detecting near-infrared fluorescence from ICG. These results suggest that DAC microendoscopy may have utility for visualizing the anatomical and, perhaps, functional changes associated with colorectal pathology for the early detection of colorectal cancer.

In vivo micro-image mosaicing.

Recent advances in optical imaging have led to the development of miniature microscopes that can be brought to the patient for visualizing tissue structures in vivo. These devices have the potential to revolutionize health care by replacing tissue biopsy with in vivo pathology. One of the primary limitations of these microscopes, however, is that the constrained field of view can make image interpretation and navigation difficult. In this paper, we show that image mosaicing can be a powerful tool for widening the field of view and creating image maps of microanatomical structures. First, we present an efficient algorithm for pairwise image mosaicing that can be implemented in real time. Then, we address two of the main challenges associated with image mosaicing in medical applications: cumulative image registration errors and scene deformation. To deal with cumulative errors, we present a global alignment algorithm that draws upon techniques commonly used in probabilistic robotics. To accommodate scene deformation, we present a local alignment algorithm that incorporates deformable surface models into the mosaicing framework. These algorithms are demonstrated on image sequences acquired in vivo with various imaging devices including a hand-held dual-axes confocal microscope, a miniature two-photon microscope, and a commercially available confocal microendoscope.

Non-invasive imaging of human embryos before embryonic genome activation predicts development to the blastocyst stage.

We report studies of preimplantation human embryo development that correlate time-lapse image analysis and gene expression profiling. By examining a large set of zygotes from in vitro fertilization (IVF), we find that success in progression to the blastocyst stage can be predicted with >93% sensitivity and specificity by measuring three dynamic, noninvasive imaging parameters by day 2 after fertilization, before embryonic genome activation (EGA). These parameters can be reliably monitored by automated image analysis, confirming that successful development follows a set of carefully orchestrated and predictable events. Moreover, we show that imaging phenotypes reflect molecular programs of the embryo and of individual blastomeres. Single-cell gene expression analysis reveals that blastomeres develop cell autonomously, with some cells advancing to EGA and others arresting. These studies indicate that success and failure in human embryo development is largely determined before EGA. Our methods and algorithms may provide an approach for early diagnosis of embryo potential in assisted reproduction.

Real-time image mosaicing for medical applications.

In this paper we describe the development of a robotically-assisted image mosaicing system for medical applications. The processing occurs in real-time due to a fast initial image alignment provided by robotic position sensing. Near-field imaging, defined by relatively large camera motion, requires translations as well as pan and tilt orientations to be measured. To capture these measurements we use 5-d.o.f. sensing along with a hand-eye calibration to account for sensor offset. This sensor-based approach speeds up the mosaicing, eliminates cumulative errors, and readily handles arbitrary camera motions. Our results have produced visually satisfactory mosaics on a dental model but can be extended to other medical images.