Artificial intelligence in the in vitro fertilization laboratory: A scoping review and SWOT-based evaluation of emerging applications
Künstliche Intelligenz im IVF-Labor: Scoping Review und SWOT-basierte Bewertung neuer Anwendungen
Clinical and Experimental Reproductive Medicine · 10.11.2026
DOI: 10.5653/cerm.2025.08900 ↗Auf einen Blick
94 Studien zeigen vor allem Anwendungen zur Embryobewertung und Implantationsprognose. Die meisten Arbeiten waren intern validiert; AUC-Werte lagen je nach Endpunkt und Datensatzgröße zwischen 0,75 und 0,92.
Fragestellung
Welche Evidenz gibt es zu KI-Anwendungen im IVF-Labor und wie ausgereift sind sie für die Translation in die Klinik?
Methode
Scoping Review nach PRISMA-ScR; PubMed, Scopus und Web of Science für Studien von Januar 2020 bis Februar 2025.
Ergebnisse
94 Studien. Häufigste Anwendungen: Embryograding und Implantationsprognose mit Convolutional Neural Networks; meist interne Cross-Validierung. AUC 0,75 bis 0,92. Spermienanalyse und Oozytenbewertung teils reproduzierbarer als manuelle Beurteilung.
Einschränkungen
Wenige Studien mit prospektiver Evaluation oder externer Validierung in multizentrischen Datensätzen.
Schlussfolgerung der Autor:innen
KI-Anwendungen im IVF-Bereich nehmen zu, sind aber meist noch in frühen Stadien der klinischen Translation. Standardisierte Datensätze, externe Validierung, interpretierbare Algorithmen und Regulierung bleiben nötig.
Kernaussagen
- Schwerpunkt auf Embryograding und Implantationsprognose
- AUC-Werte je nach Endpunkt und Datensatzgröße 0,75 bis 0,92
- Externe und prospektive Validierung selten
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Scoping Review in der Zeitschrift Clinical and Experimental Reproductive Medicine, online erschienen am 10.11.2026. Verfasst von Arefeh Akbari Javar und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Obstetrics and Gynecology, School of Medicine, Kerman University of Medical Sciences, Kerman, Iran.
Abstract (Originalsprache)
Objective
Artificial intelligence (AI) represents a potentially transformative tool in assisted reproductive technology (ART), particularly in vitro fertilization (IVF), for gamete assessment, embryo selection, non-invasive genetic testing, and laboratory automation. This scoping review synthesized recent evidence on AI applications in IVF laboratories and evaluated their translational maturity using an evidence maturity framework and a strengths, weaknesses, opportunities, and threats (SWOT) analysis.
Methods
This scoping review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Extension for Scoping Reviews guidelines. PubMed, Scopus, and Web of Science were searched for peer-reviewed studies published from January 2020 to February 2025. Eligible studies evaluated AI, machine learning, or deep learning in laboratory components of ART. Data were charted on AI methods, dataset characteristics, validation strategies, clinical endpoints, and performance metrics.
Results
Ninety-four studies were included. Most applications addressed embryo grading and implantation prediction using convolutional neural networks and were evaluated through internal cross-validation. Reported areas under the receiver operating characteristic curve ranged from 0.75 to 0.92, depending on the endpoint and dataset size. AI applications in sperm analysis and oocyte assessment showed greater reproducibility than manual evaluation. Non-invasive genetic prediction models displayed promising but heterogeneous sensitivity and remained predominantly limited to internal validation. Few studies reported prospective evaluation or external validation using multicenter datasets.
Conclusion
AI applications in IVF are expanding, but most remain at early stages of clinical translation. Standardized datasets, external validation, interpretable algorithms, and regulatory oversight remain necessary for safe, equitable clinical integration. AI should function as a decision-support tool that supplements, rather than replaces, embryologists’ expertise.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Obstetrics and Gynecology, School of Medicine, Kerman University of Medical Sciences, Kerman, Iran.
- Afzalipour Clinical Center for Infertility, Afzalipour Hospital, Kerman University of Medical Sciences, Kerman, Iran.
- Department of Biology, Faculty of Sciences, Yazd University, Yazd, Iran.
- Mother and Newborn Health Research Center, Comprehensive Research Institute for Maternal and Child Health, Shahid Sadoughi University of Medical Sciences, Yazd, Iran. dehghanpour64@yahoo.com.
Zitieren
APA
Javar, A. A., Miresmaeili, S. M., & Dehghanpour, F. (2026). Artificial intelligence in the in vitro fertilization laboratory: A scoping review and SWOT-based evaluation of emerging applications. Clinical and Experimental Reproductive Medicine. https://doi.org/10.5653/cerm.2025.08900
Vancouver
Javar AA, Miresmaeili SM, Dehghanpour F. Artificial intelligence in the in vitro fertilization laboratory: A scoping review and SWOT-based evaluation of emerging applications. Clinical and Experimental Reproductive Medicine. 2026. doi:10.5653/cerm.2025.08900
BibTeX
@article{Javar2026,
author = {Javar, Arefeh Akbari and Miresmaeili, Seyed Mohsen and Dehghanpour, Fatemeh},
title = {Artificial intelligence in the in vitro fertilization laboratory: A scoping review and SWOT-based evaluation of emerging applications},
journal = {Clinical and Experimental Reproductive Medicine},
year = {2026},
doi = {10.5653/cerm.2025.08900}
}
EUVIMED