JBRA Assist. Reprod. 2025;29(Suppl 1):60-60
POSTER PRESENTATION
doi: 10.5935/1518-0557.20250113
1Cegyr, Eugin Group - Research department (Buenos Aires, Argentina)
Objective:
• To evaluate patients’ perceptions and attitudes toward the use of AI in assessing oocyte quality during assisted reproduction treatments.
• To analyze the level of trust and acceptance among patients regarding AI-based oocyte quality assessment.
• To identify the factors influencing positive or negative perceptions of AI in assisted reproduction.
• To compare patients’ expectations of AI-based assessment with traditional evaluations performed by embryologists.
• To assess patients' perceptions of AI’s accuracy, impartiality, and clinical utility in oocyte selection.
Methods: A cross-sectional study was conducted to assess patient perceptions of using AI in oocyte quality assessment for assisted reproduction treatments between January and December 2024, in a private fertility center in Buenos Aires, Argentina. In our clinic, all the patients' oocytes are evaluated with AI, at no additional cost, and a report is delivered via email to the patient. A questionnaire was administered to all women who underwent an assisted reproduction technology cycle (including oocyte cryopreservation). The domains explored included:
• Knowledge and perception of AI
• Application of AI in reproductive medicine
• Opinion on AI in oocyte quality assessment
• Trust and safety in the use of AI for reproductive medicine
• Expectations and concerns
We described the study findings and analyzed if there was any association between the perception and any other clinical characteristics.
Results: The survey was sent to 556 women (300 undergoing in vitro fertilization and 256 undergoing oocyte freezing) with a mean age of 37.2±3.6 years. A majority (89%, 147/165) reported having at least a basic understanding of AI, 42.4% (70/165) used AI weekly, and 83% (137/165) held a predominantly positive perception of AI in medicine. Regarding oocyte quality assessment, 81% (134/165) expressed a favorable or very favorable opinion, irrespective of whether they knew their oocytes had been tested (83% vs 74%, p = 0.18). A trend was observed suggesting that positive opinions may relate to personal experience. Among those with a positive perception, key reasons included better expectation management and autonomy (35%), quick and objective access to information (21%), and receiving a written report (17%). Conversely, negative opinions were linked to perceived inaccuracy and unmet expectations (17%), lack of human involvement (13%), and unclear reasons (23%). Concerns about AI inaccuracy in oocyte quality assessment were significantly higher among women over 40 compared to younger patients (58.5% vs 39%, p = 0.03), independent of AI usage frequency. Open-ended responses revealed additional concerns, including inaccuracy (42%), lack of human interaction (43.9%), and data privacy risks (31.1%).
Conclusion: This survey revealed a highly positive perception of AI, despite limited evidence for AI algorithms in this field. Notably, participants held both positive opinions and considerable concerns about accuracy. Younger women were understandably less worried about new technologies, highlighting generational differences in attitudes towards AI in fertility care. Most respondents have a positive view of AI in reproductive medicine, but over half believe its application should be evaluated individually. Key benefits include expectation management, autonomy, quick access to objective information, and personalized egg reports. The response rate may limit this survey due to potential selection bias. Additionally, these results reflect a fertility center that integrates AI extensively, potentially leading to a more favorable patient perception of AI compared to other clinics.