EUVIMEDEUVIMED entdecken
Crossref

L26/P-200 Female body mass index is associated with differences in AI-derived oocyte score distribution: A retrospective study

O. Ammar, G. Raad, Z. Atayurt, G. Liperis, Z. Yarkiner, Y. Yilancilar, B. Yuksel, M. Serdarogullari

Human Reproduction · 01.07.2026 · 41 (Supplement_1)

DOI: 10.1093/humrep/deag083.536 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Human Reproduction, online erschienen am 08.07.2026. Verfasst von O. Ammar und 7 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Study question Is female body mass index (BMI), classified according to World Health Organization criteria, associated with differences in AI-derived oocyte classification? Summary answer Female BMI was associated with AI-derived oocyte classification, with obesity associated with fewer oocytes classified in higher categories and a greater representation of lower categories. What is known already Female obesity is a well-established determinant of oocyte competence. Obesity is associated with reduced fertilisation, impaired embryo development, and lower live birth rates. Mechanistically, excess adiposity induces metabolic dysregulation, oxidative stress, mitochondrial dysfunction, and altered granulosa cell signalling, all of which negatively affect oocyte maturation and developmental potential. Experimental and clinical studies have demonstrated obesity-related changes in follicular fluid composition, mitochondrial activity, and inflammatory pathways within the ovary. However, most evidence relies on downstream embryo or clinical outcomes, and the relationship between female BMI and objective, AI-based oocyte quality assessment at the pre-fertilisation stage remains poorly characterised. Study design, size, duration This retrospective, single-centre study analysed 2,341 metaphase II oocytes derived from 210 intracytoplasmic sperm injection (ICSI) cycles performed between March 2024 and December 2025. Female body mass index, classified according to World Health Organization criteria, was the exposure of interest, and MAGENTA AI–based oocyte scores were the outcome. All cycles used fresh metaphase II oocytes. Participants/materials, setting, methods Oocyte images were captured prior to ICSI using the MAGENTA AI system. Oocytes were categorised into four MAGENTA AI score groups: A (7.6–10.0; high), B (5.1–7.5; medium–high), C (2.6–5.0; low–medium), and D (0.0–2.5; low). Female body mass index (BMI) was classified as underweight (<18.5 kg/m²), normal weight (18.5–24.9 kg/m²), overweight (25.0–29.9 kg/m²), or obese (≥30.0 kg/m²). Main results and the role of chance MAGENTA oocyte category distribution differed across female BMI groups (Pearson χ²(9)=32.35, p = 0.00017; Cramér’s V = 0.069). The obese group showed a lower proportion of oocytes with high MAGENTA scores (Group A). Their rate was 8.4%, compared to 15.8% for the normal BMI group and 20.5% for the overweight group. Oocytes with low-medium MAGENTA scores (Group C) were most common among obese women, at 41.9%. This was higher than the normal (33.6%) and overweight (30.0%) categories. For Group B, proportions did not differ across any BMI categories. Oocytes with lower MAGENTA scores (Group D) were most frequent in underweight women (27.4%). They were also elevated in obese women (21.0%) relative to the normal (18.2%) and overweight (17.6%) groups. Post hoc comparisons showed the reduction in Group A among obese women was significant versus the overweight group (p = 0.001). The higher prevalence of Group C in obese women was significant versus the overweight (p = 0.011) and underweight (p = 0.023) groups. After age adjustment, BMI remained independently associated with MAGENTA distribution. In clustering-aware ordinal regression models, however, BMI effects were attenuated and lost statistical significance. Limitations, reasons for caution The retrospective design precludes causal inference and is susceptible to residual confounding. In addition, no clinical outcomes were evaluated, and analyses relied on categorical MAGENTA outputs, which may obscure more subtle variation in oocyte quality. Wider implications of the findings Female obesity was associated with AI-based shifts towards lower oocyte quality prior to fertilisation. BMI-stratified AI oocyte scoring may support early cycle prognostication and patient counselling, and provide an objective framework to monitor the impact of preconception weight optimisation before downstream embryo and clinical outcomes. Trial registration number No

Zitieren

APA

Ammar, O., Raad, G., Atayurt, Z., Liperis, G., Yarkiner, Z., Yilancilar, Y., Yuksel, B., & Serdarogullari, M. (2026). L26/P-200 Female body mass index is associated with differences in AI-derived oocyte score distribution: A retrospective study. Human Reproduction, 41(Supplement_1). https://doi.org/10.1093/humrep/deag083.536

Vancouver

Ammar O, Raad G, Atayurt Z, Liperis G, Yarkiner Z, Yilancilar Y, et al. L26/P-200 Female body mass index is associated with differences in AI-derived oocyte score distribution: A retrospective study. Human Reproduction. 2026;41(Supplement_1). doi:10.1093/humrep/deag083.536

BibTeX
@article{Ammar2026,
  author = {Ammar, O. and Raad, G. and Atayurt, Z. and Liperis, G. and Yarkiner, Z. and Yilancilar, Y. and Yuksel, B. and Serdarogullari, M.},
  title = {L26/P-200 Female body mass index is associated with differences in AI-derived oocyte score distribution: A retrospective study},
  journal = {Human Reproduction},
  year = {2026},
  volume = {41},
  number = {Supplement_1},
  doi = {10.1093/humrep/deag083.536}
}
Kostenlos starten →