The effect of early irregular cell division of human embryos on blastocyst euploidy: considerations from the subsequent development of the blastomeres by direct or reverse cleavage.

OBJECTIVE: To investigate whether blastocysts that divide irregularly reduce subsequent blastocyst euploidy. DESIGN: Retrospective study. SETTING: Private clinic. PATIENT(S): A total of 122 blastocysts for which consent for disposal and research use was obtained. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Results of next-generation sequencing analysis of the blastocysts and whether blastomeres by normal or irregular divisions subsequently participated in blastocyst formation or not. RESULT(S): The embryos were classified according to their dynamics until the second cleavage. The blastocyst euploidy rates were 33.3% (19/57) in the normal cleavage (NC) group, 38.3% (18/47) in the direct cleavage (embryos with one cell dividing into 3 cells) (DC) group, and 72.2% (13/18) in the reverse cleavage (RC) (embryos with fused cells once divided) group. The rate of the RC group was significantly higher than that of the NC group. The blastocyst participation rate of the blastomeres were 95.6% in the NC group and 56.5% in that derived from DC of the first cleavage, and 91.7% in that of blastomeres derived from normal division of the second cleavage and 53.6% in that derived from DC of the second cleavage, both of which were significantly lower in the latter. In the RC group, the rates of fused and nonfused blastomeres were 62.1% and 87.5%, respectively, with no significant difference. CONCLUSION(S): The blastomeres generated by DC were often excluded from blastocyst formation, and we speculate that this is one reason why their division does not reduce blastocyst euploidy. The association between RC and euploidy of blastocysts merits further study.

Evaluation of artificial intelligence using time-lapse images of IVF embryos to predict live birth.

RESEARCH QUESTION: Can artificial intelligence (AI) improve the prediction of live births based on embryo images? DESIGN: The AI system was created by using the Attention Branch Network associated with deep learning to predict the probability of live birth from 141,444 images recorded by time-lapse imaging of 470 transferred embryos, of which 91 resulted in live birth and 379 resulted in non-live birth that included implantation failure, biochemical pregnancy and clinical miscarriage. The possibility that the calculated confidence scores of each embryo and the focused areas visualized in each embryo image can help predict subsequent live birth was examined. RESULTS: The AI system for the first time successfully visualized embryo features in focused areas that had potential to distinguish between live and non-live births. No visual feature of embryos were visualized that were associated with live or non-live births, although there were many images in which high-focused areas existed around the zona pellucida. When a cut-off level for the confidence score was set at 0.341, the live birth rate was significantly greater for embryos with a score higher than the cut-off level than for those with a score lower than the cut-off level (P < 0.001). In addition, the live birth rate of embryos with good morphological quality and confidence scores higher than 0.341 was 41.1%. CONCLUSIONS: The authors have created an AI system with a confidence score that is useful for non-invasive selection of embryos that could result in live birth. Further study is necessary to improve selection accuracy.