L26/P-056 Comparison of embryo morphokinetics after Intra Cytoplasmic Sperm Injection (ICSI) with an AI-enhanced and traditional sperm selection method under routine laboratory conditions
Human Reproduction · 01.07.2026 · 41 (Supplement_1)
DOI: 10.1093/humrep/deag083.392 ↗Kurz gefasst
Fachartikel in der Zeitschrift Human Reproduction, online erschienen am 08.07.2026. Verfasst von A. Agathangelou und 4 weiteren Autor:innen.
Abstract (Originalsprache)
Study question Can we improve the outcome of ICSI by utilising the Sperm Tracer AI tool to select optimal sperm for ICSI in autologous treatment cycles? Summary answer AI-driven sperm selection was associated with statistically significantly higher blastocyst formation, with trends toward higher IDA scores and earlier cleavage using the Sperm Tracer system. What is known already The traditional way of selecting sperm for ICSI (Intracytoplasmic Sperm Injection) by embryologists based on morphological evaluation is prone to subjective variability, which may negatively affect fertilisation and embryo development. Artificial Intelligence (AI) platforms provide more reliable (standardised and objective) methods of evaluating sperm as selection is based on morphological and motility-related parameters, resulting in improved embryo quality. However, the full potential of these AI tools to enhance ІCSI success will only be realised when their usefulness in improving ICSI outcome can be established through extensive clinical validation, with meaningful post ICSI embryo developmental results. Study design, size, duration Between 2025–2026, 224 sibling MII oocytes from 22 autologous ICSI cycles were randomly divided where feasible, into two groups and inseminated by traditional ICSI (sperm selected by embryologists) (n = 166) or AI-assisted sperm selection (sperm selected by Sperm Tracer AI tool) (n = 58). The oocytes were cultured and monitored up to the blastocyst stage. This matched design enabled within-patient comparisons on the fertilisation and blastocyst yield rates, which were assessed using appropriate statistical analyses for paired samples. Participants/materials, setting, methods Sperm samples were prepared using Zymot microfluidic devices. Sperm Tracer AI sperm selection was based on morphological and kinetic parameters including head, mid-piece tail (VCL, VSL, VAP, ALH, LIN, STR, WOB, DNC). Both, benchtop (Planner) and timelapse incubator (Embryoscope+) were used to culture embryos. Embryo morphokinetics and quality were assessed using the IDAScore system in conjunction with standard embryologist evaluation. Statistical analysis included Wilcoxon signed-rank tests for pair-wise comparison and McNemar’s tests for binary outcomes Main results and the role of chance Spermatozoa selected for ICSI with the Sperm Tracer (ST-AI) system significantly increased blastocyst formation rates (59.7% vs 55.2%, p = 0.015). Notable trends were observed with higher rates of good quality blastocysts (34.1% vs 23.7% - relative difference: +43.9%), higher implantation potential with IDA scores (4.23 vs 2.96, p = 0.064) and earlier cleavage (27.4h vs 29.3h, p = 0.074) in the same ST-AI group. IDA scores correlated more strongly with cleavage timing in ST-AI embryos (ρ=-0.36, p = 0.016) than controls (ρ=-0.12, p = 0.211), suggesting more coordinated development. No significant differences were found in fertilisation rates (ST-AI: 80.5% vs control:73.5%, p = 0.201) or proportions of top-quality blastocysts (p = 1.000). Notably, while conventional selection showed statistically significant age-related decline in blastocyst rates (ρ=-0.559, p = 0.016), this relationship was absent with ST-AI (ρ=-0.181, p = 0.471). Patients ≥35 years experienced 66.5% relative improvement with ST-AI versus 43.3% for younger patients, suggesting AI may attenuate age-related declines in embryo developmental competence. Taken together, the results suggest that Sperm Tracer AI achieves comparable efficiency to conventional sperm selection while generating a hypothesis that AI-assisted selection may attenuate age-related declines in embryo developmental competence. The age effect observed with ST-AI warrants further investigation in larger trials to validate clinical benefits, particularly for older patient populations. Limitations, reasons for caution The study’s limited sample size and monocentric design may restrict statistical power, generalisability and power for secondary outcomes. Findings should be interpreted with caution, as they may not be applicable to broader populations.Future multi-center studies with larger sample sizes will be needed to confirm results and their clinical implications. Wider implications of the findings Sperm Tracer-AI improved blastocyst formation and suggested attenuation of age-related effects in this pilot study. Confirmation could optimise ICSI, particularly benefiting patients ≥35 years who showed 66% relative improvement. AI offers standardised sperm selection, potentially increasing embryo utilisation and treatment success while addressing age-related infertility challenges. Trial registration number No
Zitieren
APA
Agathangelou, A., Stylianou, C. A., Mantas, D., Kyriakopoulou, E., & Mavrides, A. (2026). L26/P-056 Comparison of embryo morphokinetics after Intra Cytoplasmic Sperm Injection (ICSI) with an AI-enhanced and traditional sperm selection method under routine laboratory conditions. Human Reproduction, 41(Supplement_1). https://doi.org/10.1093/humrep/deag083.392
Vancouver
Agathangelou A, Stylianou CA, Mantas D, Kyriakopoulou E, Mavrides A. L26/P-056 Comparison of embryo morphokinetics after Intra Cytoplasmic Sperm Injection (ICSI) with an AI-enhanced and traditional sperm selection method under routine laboratory conditions. Human Reproduction. 2026;41(Supplement_1). doi:10.1093/humrep/deag083.392
BibTeX
@article{Agathangelou2026,
author = {Agathangelou, A. and Stylianou, C. Antoniadou and Mantas, D. and Kyriakopoulou, E. and Mavrides, A.},
title = {L26/P-056 Comparison of embryo morphokinetics after Intra Cytoplasmic Sperm Injection (ICSI) with an AI-enhanced and traditional sperm selection method under routine laboratory conditions},
journal = {Human Reproduction},
year = {2026},
volume = {41},
number = {Supplement_1},
doi = {10.1093/humrep/deag083.392}
}
EUVIMED