O-064 Artificial intelligence in embryo selection of IVF
Abstract Abstract text Some studies have discussed the use of artificial intelligence and machine learning in the assessment and selection of embryos for in vitro fertilization. Complete artificial intelligence acquired using CNN’s dark box algorithm could be highly useful in assessing in embryos, though it could be difficult to perform the external validation necessary to confirm its value. But due to the inherent drawbacks in complete artificial intelligence assessing in vitro developmental embryos, such as lacking results of discard embryos, dislocations between computer scientist and embryologist, low explanatory values in dark box algorithm, here, we suggest training computers to recognize the target region (internal pellucid zone region) and the features of embryos, then continuously score the embryos starting at in vitro fertilization through the zygote to the blastocyst stage. Parameters suitable for use with various endpoints in treatment sequence could be found by AI. Further clinical studies should be performed to validate the parameters and AI needed. Trial registration number: Study funding: Funding source: