eduzhai > Applied Sciences > Engineering >

Deep learning mediated single time-point image-based prediction of embryo developmental outcome at the cleavage stage

  • Save

... pages left unread,continue reading

Document pages: 31 pages

Abstract: In conventional clinical in-vitro fertilization practices embryos aretransferred either at the cleavage or blastocyst stages of development.Cleavage stage transfers, particularly, are beneficial for patients withrelatively poor prognosis and at fertility centers in resource-limited settingswhere there is a higher chance of developmental failure in embryos in-vitro.However, one of the major limitations of embryo selections at the cleavagestage is the availability of very low number of manually discernable featuresto predict developmental outcomes. Although, time-lapse imaging systems havebeen proposed as possible solutions, they are cost-prohibitive and requirebulky and expensive hardware, and labor-intensive. Advances in convolutionalneural networks (CNNs) have been utilized to provide accurate classificationsacross many medical and non-medical object categories. Here, we report anautomated system for classification and selection of human embryos at thecleavage stage using a trained CNN combined with a genetic algorithm. Thesystem selected the cleavage stage embryo at 70 hours post insemination (hpi)that ultimately developed into top-quality blastocyst at 70 hpi with 64 accuracy, outperforming the abilities of embryologists in identifying embryoswith the highest developmental potential. Such systems can have a significantimpact on IVF procedures by empowering embryologists for accurate andconsistent embryo assessment in both resource-poor and resource-rich settings.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×