Artificial Intelligence and Infertility: A Systematic Review

Systematic review maps AI applications in male fertility — sperm analysis, treatment prediction — but reports no pooled numbers

Journal: The French Journal of Urology | Published: 2026-03-03 | Type: Systematic Review | PMID: 41786088 Authors: Al Mahdi B, El Kaddoury H, Tollet V, Mjaess G, Roumeguère T — all from the Department of Urology, Hôpital Universitaire de Bruxelles (Université Libre de Bruxelles) Funding/COI: Funding not reported. Authors declare no competing interests.

Summary

This systematic review surveys how AI tools are being applied to male infertility — primarily automated sperm analysis (morphology, motility, DNA fragmentation detection) and predictive models for treatment outcomes. The abstract makes no reference to how many studies were included, what their sample sizes were, or what effect sizes AI systems achieved versus conventional analysis. What reads like a research review reads more like an advocacy position paper.

Claims

• AI enables "more accurate sperm assessments" compared to conventional methods — no accuracy figures provided
• Predictive modeling is described as enabling "individualized treatment strategies" — no success-rate data cited
• Integration raises ethical concerns around data privacy and automation in clinical decision-making — not quantified or cited
• AI is characterized as "poised to transform" andrological care — speculative, no evidence cited

Study Quality

This is a systematic review published without a disclosed protocol or registration number (no PROSPERO ID mentioned), with no inclusion/exclusion criteria, no reported number of included studies, no quality assessment tool (GRADE, AMSTAR, ROBIS), and no pooled data or effect estimates. That is not a systematic review in any meaningful methodological sense — it is a narrative survey wearing a systematic review label. The absence of a PRISMA flow diagram, study count, or any quantitative synthesis makes it impossible to assess what evidence base was actually reviewed.

The abstract is a string of aspirational statements ("reshaping," "transforming," "poised to transform") without a single number. For a field where the primary question is how much better AI sperm analysis performs than a trained andrologist, the absence of any comparative accuracy data is a serious gap.

Red Flags

• No study count, no PRISMA flow, no quality appraisal — the three things that make a systematic review a systematic review
• No funding source disclosed for a paper from a major academic medical center
• Abstract is entirely qualitative despite reviewing a literature full of quantitative performance metrics
• All five authors share the same single department — no methodologist, no external perspective
• "Poised to transform" and "reshaping paradigms" language signals advocacy over analysis
• Published in a journal (French Journal of Urology) that has not historically been a venue for high-impact AI methodology work

Strengths

• COI disclosure is clean
• Topic is clinically relevant — automated semen analysis is an active and underdeveloped area
• Ethical discussion of data privacy and clinician oversight is a necessary counterweight to AI hype in this space

Verdict

The title says systematic review; the abstract delivers a brochure. Without a disclosed protocol, study count, quality assessment, or a single quantitative finding, this paper cannot function as a reliable map of the AI-in-infertility evidence base. It may be useful as a starting-point reading list if the full text contains a proper study table — but on the evidence of the abstract alone, treat it as a narrative opinion piece. The topic deserves a rigorous systematic review; this does not appear to be one.