Artificial intelligence-based embryo ranking can match or improve traditional assessments but cannot predict an aneuploid chromosomal constitution
Rattachement africain : it, us, nl. Niveau de preuve : code pays fourni par la source.
Le résumé fourni par la source
OBJECTIVE: To evaluate whether artificial intelligence (AI)-based blastocyst ranking tools can outperform traditional noninvasive assessments in prioritizing euploid embryos for transfer. DESIGN: Retrospective single-center cohort study. SUBJECTS: The study included 786 preimplantation genetic testing for aneuploidy cycles performed between 2013 and 2020 after embryo culture in a time-lapse incubator (TLI). The analysis focused on 279 cycles producing ≥3 blastocysts, of which ≥1 was euploid and ≥1 aneuploid embryo. EXPOSURE: ); and two versions of a commercially-available AI model. The coefficient of variation (CV = SD/mean) within each cohort of blastocysts was then calculated to estimate the morphological variability perceived by each method among sibling embryos, and its association with ranking accuracy was investigated. MAIN OUTCOME MEASURES: Accuracy in prioritizing euploid blastocysts within cohorts of sibling embryos, assessed by the proportion of cycles in which a euploid embryo was ranked first, and association between intracohort CV and ranking performance. RESULTS: The AI model version-2 showed the highest rate of correctly prioritized euploid blastocysts (68%), followed by morphodynamic assessment (64%), AI model version-1 (62%), Gardner's score (58%), and random guessing (44%). Larger intracohort CVs for Gardner's scores were positively associated with prioritization of euploid embryos. No such associations were observed with AI-based CVs. CONCLUSION: The AI-powered tools can match or slightly improve traditional noninvasive assessments. However, no blastocyst ranking strategy on the basis of morphology/morphodynamics may prevent aneuploid embryos from being prioritized, even in the presence of euploid siblings.
Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.
Le contrôle bibliographique ouvert
DOI retrouvé dans Crossref DOI retrouvé ; titre concordant.
- Titre Crossref
- Artificial intelligence-based embryo ranking can match or improve traditional assessments but cannot predict an aneuploid chromosomal constitution
- Date Crossref
- 01/08/2026
- Éditeur
- Elsevier BV
- Type
- journal-article
Ce recoupement confirme des métadonnées liées au DOI. Il ne confirme ni la méthode ni les conclusions de l’étude, et il ne compte pas comme une seconde source scientifique indépendante.
Les institutions déclarées
Une affiliation ne permet pas de déduire la nationalité d’un auteur.