EUVIMEDEUVIMED entdecken
Crossref

L26/P-263 Embryo Ranking Intelligent Classification Assistant (ERiCA) predicts live birth: what biological features underlie this prediction?

A. Ito, Y. Katagiri, M. Nakata, H. Kobayashi

Human Reproduction · 01.07.2026 · 41 (Supplement_1)

DOI: 10.1093/humrep/deag083.599 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Human Reproduction, online erschienen am 08.07.2026. Verfasst von A. Ito und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Study question How accurately does the ERiCA score predict live birth, and how is it associated with embryonic morphological and dynamic indicators? Summary answer The ERiCA score predicted live birth with a ROC-AUC of 0.66. Stepwise differences were observed across ERiCA categories in live birth, morphology, and post-warming re-expansion. What is known already Embryo selection is central to improving live birth rates in assisted reproduction. Recently, artificial intelligence–based embryo assessment tools have been introduced to estimate embryo competence beyond conventional morphology. ERiCA is an AI-derived score developed to predict embryonic euploidy; however, its predictive performance for live birth and the biological embryo characteristics underlying this prediction have not been fully elucidated. Study design, size, duration This cohort study included 522 hormone replacement frozen–thawed single blastocyst transfer cycles performed at our centre between 2022 and 2022. The primary outcome was live birth, and the predictive performance of ERiCA was evaluated using ROC analysis. Secondary analyses compared live birth rates, the proportion of good-quality blastocysts, and post-warming re-expansion rates across four ERiCA categories (poor, fair, good, optimal). Participants/materials, setting, methods ERiCA scores were obtained from blastocyst images acquired before freezing. Blastocyst morphology was assessed using the Gardner classification, with good-quality blastocysts defined as ≥ 3BB. Post-warming re-expansion was defined as an embryo area ≥90.2% of the inner zona pellucida area at 10 minutes after warming, based on our previously published cutoff. Differences across ERiCA categories were analysed using chi-square tests, with assessment of stepwise trends. Main results and the role of chance ERiCA predicted live birth with an AUC of 0.661 (95% CI 0.606–0.717), indicating moderate discrimination. Live birth rates increased stepwise across ERiCA categories, from 7.1% in the poor group to 14.2% in the fair group, 27.6% in the good group, and 41.3% in the optimal group (p < 0.001). Post-warming embryo re-expansion also differed significantly across ERiCA categories, with rates of 50.0%, 47.0%, 58.4%, and 76.0% in the poor, fair, good, and optimal groups, respectively (p < 0.001). In contrast, the proportion of good-quality blastocysts did not show a consistent stepwise increase across ERiCA categories (92.9%, 77.6%, 89.3%, and 94.7%; χ² p < 0.001). These findings suggest that ERiCA classification aligns more strongly with dynamic embryo competence than with morphology alone, despite only moderate predictive accuracy for live birth. Limitations, reasons for caution This single-centre observational study lacked external validation. Laboratory-specific culture conditions and category thresholds may limit generalisability, and residual confounding cannot be excluded. Wider implications of the findings Although ERiCA shows only moderate predictive performance for live birth, it represents an interpretable AI-based score that reflects embryo dynamic competence rather than morphology alone. Trial registration number No

Zitieren

APA

Ito, A., Katagiri, Y., Nakata, M., & Kobayashi, H. (2026). L26/P-263 Embryo Ranking Intelligent Classification Assistant (ERiCA) predicts live birth: what biological features underlie this prediction?. Human Reproduction, 41(Supplement_1). https://doi.org/10.1093/humrep/deag083.599

Vancouver

Ito A, Katagiri Y, Nakata M, Kobayashi H. L26/P-263 Embryo Ranking Intelligent Classification Assistant (ERiCA) predicts live birth: what biological features underlie this prediction?. Human Reproduction. 2026;41(Supplement_1). doi:10.1093/humrep/deag083.599

BibTeX
@article{Ito2026,
  author = {Ito, A. and Katagiri, Y. and Nakata, M. and Kobayashi, H.},
  title = {L26/P-263 Embryo Ranking Intelligent Classification Assistant (ERiCA) predicts live birth: what biological features underlie this prediction?},
  journal = {Human Reproduction},
  year = {2026},
  volume = {41},
  number = {Supplement_1},
  doi = {10.1093/humrep/deag083.599}
}
Kostenlos starten →