چکیده مقاله
Background: Embryo selection is critical to IVF success Traditional morphological assessment is limited by subjectivity AI and deep learning offer a paradigm shift towards objective, data driven embryo viability analysis Objective: To provide a critical synthesis of the application of Artificial Intelligence AI in embryo selection for in vitro fertilization IVF , focusing on its evolution, clinical evidence, and implementation challenges Methods: A narrative review and critical appraisal of the literature was conducted, examining the technological foundations of AI, its comparative performance against embryologists, and the transition from retrospective validation to prospective clinical trials Results: AI models demonstrate superior or non inferior performance compared to embryologists in predicting key developmental outcomes Major challenges to clinical adoption include the black box problem, data bias affecting generalizability, and evolving regulatory frameworks The role of the embryologist is consequently evolving towards that of an AI augmented curator Conclusion: AI is a validated and transformative tool in modern embryology Its responsible integration into mainstream IVF practice requires robust clinical validation through multi center trials, the development of explainable AI, and clear ethical guidelines to ensure it augments clinical expertise
کلیدواژهها
نویسندگان
Sedigheh Ekraminasab
Department of Hematology and Blood Banking, School of Allied Medical Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
شیوه ارجاع
Ekraminasab, Sedigheh,1404,Artificial Intelligence in Embryo Selection: A New Paradigm for Assisted Reproduction,17th International Conference on New Findings in Midwifery, Women, Maternity and Infertility
ارائهشده در
مجموعه مقالات هفدهمین کنفرانس بین المللی یافته های نو در مامایی،زنان ،زایمان و نازایی7 آذر 1404