Artificial intelligence in assisted reproductive technology: separating the dream from reality

Résumé fourni par la source

This paper critically reviews the role of artificial intelligence (AI) in assisted reproductive technology (ART), a nascent field that has emerged over the last decade. While AI holds immense promise for enhancing IVF efficiency, standardization, and outcomes, its current trajectory reveals significant challenges. Much of the recent literature presents variations on established methodologies rather than groundbreaking advancements, with many studies lacking clear clinical applications or outcome-driven validations. Moreover, the growing enthusiasm for AI in ART is often accompanied by undue hype that obscures its realistic potential and fosters inflated expectations. Despite these limitations, AI-driven innovations such as advanced image analysis, personalized protocols, and automation of embryology workflows are beginning to show value. Machine learning algorithms and robotics may help address inefficiencies, alleviate staff shortages, and improve decision-making in the IVF laboratory. However, progress is tempered by drawbacks including ethical concerns, limited transparency in AI systems, and regulatory impediments. Data-sharing barriers in our field hinder AI tool development significantly. Energy-intensive computational processes and expanding data centers also raise sustainability concerns, underscoring the need for environmentally responsible development. As the field evolves, it must emphasize rigorous validation, collaborative data frameworks, and alignment with the needs of ART practitioners and patients.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.