L26/P-242 Artificial intelligence-assisted versus manual sperm selection in ICSI cycles: a prospective comparative study
Human Reproduction · 01.07.2026 · 41 (Supplement_1)
DOI: 10.1093/humrep/deag083.578 ↗Kurz gefasst
Fachartikel in der Zeitschrift Human Reproduction, online erschienen am 08.07.2026. Verfasst von S. Zakarneh und 6 weiteren Autor:innen.
Abstract (Originalsprache)
Study question Does AI-assisted sperm selection during ICSI yield fertilization and embryo development outcomes comparable to conventional manual sperm selection performed by an expert embryologist? Summary answer AI-assisted sperm selection achieved fertilization and embryo development outcomes comparable to expert selection, supporting its potential role as a standardization and decision-support tool in ICSI. What is known already AI-assisted approaches are increasingly incorporated into assisted reproductive technology to enhance objectivity and reduce operator-dependent variability. Conventional manual sperm selection during ICSI relies on subjective embryologist assessment of motility and morphology and is influenced by individual experience and performance. AI-based sperm selection systems, such as the Sperm Identification Device (SiD; IVF2.0 Ltd), apply real-time image analysis to rank individual spermatozoa based on complex motility patterns. Previous studies have reported improved or comparable fertilization and embryo development outcomes with SiD-assisted selection versus conventional manual selection; however, external validation across different patient populations and clinical settings remains essential. Study design, size, duration This was a single-centre prospective observational study conducted between May 2025 and January 2026, including 61 ICSI cycles and 894 injected oocytes. In three cycles, all oocytes were injected using sperm selected by SiD. In the remaining cycles, oocytes were randomly allocated to injection with sperm selected either by SiD or by conventional manual selection performed by a single senior embryologist with over 10 years of experience. Participants/materials, setting, methods Patients undergoing ICSI were included. Semen samples with a concentration >1 million/mL and normal total motility were eligible. Sperm samples were prepared using a swim-up technique. In split cycles, oocytes were randomly assigned to injection with sperm selected by either SiD-assisted or conventional manual selection. All ICSI procedures and embryological assessments were performed by the same senior embryologist. Outcomes were analysed using statistical methods, including mixed-effects models to account for clustering at the patient level Main results and the role of chance A total of 459 oocytes were injected using conventional manual sperm selection and 435 using SiD-assisted selection. Fertilization rates were comparable between groups (manual: 79.3% ± 18.8% vs. SiD: 79.3% ± 17.5%). Day-2 embryo development rates were 94.5% ± 12.5% in the manual group and 91.9% ± 17.2% in the SiD group. Blastocyst formation rates were 53.6% ± 26.5% following manual selection and 51.0% ± 27.0% following SiD-assisted selection. Mixed-effects logistic regression models showed no statistically significant differences between manual and SiD-assisted sperm selection for fertilization (OR 0.99, 95% CI 0.72–1.38; p = 0.97), day-2 embryo development (OR 0.70, 95% CI 0.36–1.34; p = 0.28), or blastocyst formation (OR 0.91, 95% CI 0.66–1.26; p = 0.58). These findings suggest that SiD achieves embryological outcomes comparable to expert manual selection, supporting its potential role as a standardized, decision-support tool without introducing systematic bias. Limitations, reasons for caution This study was conducted at a single center with a moderate sample size. All procedures were performed by a senior embryologist; evaluation with less experienced embryologists may be valuable to assess workflow applicability. SiD primarily evaluates sperm motility, while final sperm selection still requires an embryologist’s confirmation of morphology assessment. Wider implications of the findings Not all laboratories have consistent access to highly experienced embryologists. In this study, no statistically significant differences were observed between AI-assisted and senior manual sperm selection across key embryological outcomes. These findings suggest AI-assisted approaches may help support workflow standardisation, warranting prospective evaluation as clinical experience with such systems increases. Trial registration number No
Zitieren
APA
Zakarneh, S., Ibraheem, M., Seder, M., AbuMuferreh, Y., Abukhaizaran, Z., Abukhayzaran, N., & Abukhaizaran, R. (2026). L26/P-242 Artificial intelligence-assisted versus manual sperm selection in ICSI cycles: a prospective comparative study. Human Reproduction, 41(Supplement_1). https://doi.org/10.1093/humrep/deag083.578
Vancouver
Zakarneh S, Ibraheem M, Seder M, AbuMuferreh Y, Abukhaizaran Z, Abukhayzaran N, et al. L26/P-242 Artificial intelligence-assisted versus manual sperm selection in ICSI cycles: a prospective comparative study. Human Reproduction. 2026;41(Supplement_1). doi:10.1093/humrep/deag083.578
BibTeX
@article{Zakarneh2026,
author = {Zakarneh, S. and Ibraheem, M. and Seder, M. and AbuMuferreh, Y. and Abukhaizaran, Z. and Abukhayzaran, N. and Abukhaizaran, R.},
title = {L26/P-242 Artificial intelligence-assisted versus manual sperm selection in ICSI cycles: a prospective comparative study},
journal = {Human Reproduction},
year = {2026},
volume = {41},
number = {Supplement_1},
doi = {10.1093/humrep/deag083.578}
}
EUVIMED