Artificial Intelligence in the IVF Laboratory: A Systematic Review and SWOT-Based Evaluation of Emerging Applications

Background: Artificial intelligence (AI) has rapidly emerged as a transformative force in assisted reproductive technologies (ART), improving precision, objectivity, and reproducibility in laboratory workflows. From sperm and oocyte assessment to embryo grading and non-invasive genetic testing, AI-driven systems are redefining the embryology laboratory environment. Objective: This review aims to systematically evaluate recent applications of AI in ART laboratories, identify methodological strengths and limitations, and provide a comprehensive SWOT-based analysis to guide future research and implementation. Methods: A systematic search of PubMed, Scopus, and Web of Science databases was performed for studies published between ۲۰۲۰ and ۲۰۲۵. Inclusion criteria focused on original research and reviews investigating AI, machine learning (ML), or deep learning (DL) within laboratory aspects of ART. Extracted data were categorized by application area, including sperm analysis, oocyte evaluation, embryo viability prediction, non-invasive diagnostics, and laboratory automation. Results: A total of ۹۴ eligible studies were analyzed. Most employed DL and convolutional neural network (CNN) models for image-based assessment, achieving up to ۹۷% accuracy in gamete and embryo evaluation. Approximately ۲۵% integrated time-lapse imaging, and ۱۵% combined AI with multi-omics or cfDNA-based diagnostics. The SWOT analysis revealed key strengths (accuracy, reproducibility, predictive power), weaknesses (data heterogeneity, cost, ethical concerns), opportunities (automation, personalized medicine, integration with robotics), and threats (data privacy, bias, regulatory gaps). Conclusions: AI is not a replacement for human expertise but a powerful ally that enhances decision-making in ART laboratories. Standardized datasets, explainable algorithms, and ethical frameworks are essential for ensuring transparent, equitable, and clinically validated implementation of AI in reproductive medicine.

Keywords:

Artificial intelligence , Machine learning , Assisted reproductive technology , Embryology , IVF laboratory , Deep learning , Non-invasive embryo testing