Deep learning for embryo evaluation using time-lapse: a systematic review of diagnostic test accuracy.

OBJECTIVE: This study aimed to investigate the accuracy of convolutional neural network models in the assessment of embryos using time-lapse monitoring. DATA SOURCES: A systematic search was conducted in PubMed and Web of Science databases from January 2016 to December 2022. The search strategy was carried out by using key words and MeSH (Medical Subject Headings) terms. STUDY ELIGIBILITY CRITERIA: Studies were included if they reported the accuracy of convolutional neural network models for embryo evaluation using time-lapse monitoring. The review was registered with PROSPERO (International Prospective Register of Systematic Reviews; identification number CRD42021275916). METHODS: Two reviewer authors independently screened results using the Covidence systematic review software. The full-text articles were reviewed when studies met the inclusion criteria or in any uncertainty. Nonconsensus was resolved by a third reviewer. Risk of bias and applicability were evaluated using the QUADAS-2 tool and the modified Joanna Briggs Institute or JBI checklist. RESULTS: Following a systematic search of the literature, 22 studies were identified as eligible for inclusion. All studies were retrospective. A total of 522,516 images of 222,998 embryos were analyzed. Three main outcomes were evaluated: successful in vitro fertilization, blastocyst stage classification, and blastocyst quality. Most studies reported >80% accuracy, and embryologists were outperformed in some. Ten studies had a high risk of bias, mostly because of patient bias. CONCLUSION: The application of artificial intelligence in time-lapse monitoring has the potential to provide more efficient, accurate, and objective embryo evaluation. Models that examined blastocyst stage classification showed the best predictions. Models that predicted live birth had a low risk of bias, used the largest databases, and had external validation, which heightens their relevance to clinical application. Our systematic review is limited by the high heterogeneity among the included studies. Researchers should share databases and standardize reporting.

Previous infection with SARS-CoV-2 impacts embryo morphokinetics but not clinical outcomes in a time-lapse imaging system.

The goal for the present study was to investigate whether previous infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) may compromise embryo morphokinetics and implantation. For that, a historical cohort study was performed in a private university-affiliated in vitro fertilization center. The study included 1628 embryos from 88 patients undergoing intracytoplasmic sperm injection (ICSI) cycles. Patients were age-matched in a 1:3 ratio to either a coronavirus disease (COVID) group, including patients with a positive SARS-CoV-2 immunoglobulin test (n = 22 patients, 386 embryos), or a control group, including patients with a negative SARS-CoV-2 immunoglobulin test (n = 66, 1242 embryos). The effect of previous infection with SARS-CoV-2 on morphokinetic events and ICSI outcomes was evaluated. Embryos derived from patients in the COVID group presented longer time to pronuclei appearance and fading, time to form two, three, four and five cells, and time to blastulation. The durations of the third cell cycle and to time to complete synchronous divisions were also significantly increased in the COVID group compared with the control group, whereas known implantation diagnosis score Day 5 ranked significantly lower in the COVID group. No differences were observed between the COVID and control groups on clinical outcomes. In conclusion, patients planning parenthood, who have recovered from COVID-19 infection, must be aware of a possible effect of the infection on embryo development potential.

Synthesis and Anticancer and Antiviral Activities of C-2'-Branched Arabinonucleosides.

d-Arabinofuranosyl-pyrimidine and -purine nucleoside analogues containing alkylthio-, acetylthio- or 1-thiosugar substituents at the C2' position were prepared from the corresponding 3',5'-O-silylene acetal-protected nucleoside 2'-exomethylenes by photoinitiated, radical-mediated hydrothiolation reactions. Although the stereochemical outcome of the hydrothiolation depended on the structure of both the thiol and the furanoside aglycone, in general, high d-arabino selectivity was obtained. The cytotoxic effect of the arabinonucleosides was studied on tumorous SCC (mouse squamous cell) and immortalized control HaCaT (human keratinocyte) cell lines by MTT assay. Three pyrimidine nucleosides containing C2'-butylsulfanylmethyl or -acetylthiomethyl groups showed promising cytotoxicity at low micromolar concentrations with good selectivity towards tumor cells. SAR analysis using a methyl ß-d-arabinofuranoside reference compound showed that the silyl-protecting group, the nucleobase and the corresponding C2' substituent are crucial for the cell growth inhibitory activity. The effects of the three most active nucleoside analogues on parameters indicative of cytotoxicity, such as cell size, division time and cell generation time, were investigated by near-infrared live cell imaging, which showed that the 2'-acetylthiomethyluridine derivative induced the most significant functional and morphological changes. Some nucleoside analogues also exerted anti-SARS-CoV-2 and/or anti-HCoV-229E activity with low micromolar EC50 values; however, the antiviral activity was always accompanied by significant cytotoxicity.

Cytotoxic T lymphocytes targeting a conserved SARS-CoV-2 spike epitope are efficient serial killers.

Understanding immune response to infections and vaccines lags understanding humoral responses. While neutralizing antibody responses wane over time, T cells are instrumental in long-term immunity. We apply machine learning and time-lapse imaging microscopy in nanowell grids (TIMING) to study thousands of videos of T cells with specificity for SARS-CoV-2 eliminating targets bearing spike protein as a surrogate for viral infection. The data on effector functions, including cytokine secretion and cytotoxicity, provide the first direct evidence that cytotoxic T lymphocytes from a convalescent patient targeting an epitope conserved across all known variants of concern are serial killers capable of eliminating multiple infected target cells. These data have implications for vaccine development and for the recovery and monitoring of infected individuals.