Abstract:

Artificial intelligence (AI) has been applied extensively across the whole spectrum of assisted reproduction, covering female infertility diagnosis, follicular monitoring, embryo selection, pregnancy outcome prediction, male infertility assessment, follow-up management, patient education and psychological support. Existing studies have demonstrated that AI can improve clinical efficiency, streamline information administration and reduce the workload of medical staff, and that AI has certain strengths in the patient health education and follow-up quality control. Although the patients undergoing assisted reproduction treatment generally recognize the auxiliary value of AI, they prefer to regard it merely as a supporting tool for physicians' clinical judgements and nursing services, rather than a complete replacement for medical practitioners. Patients' acceptance of AI is influenced by the demographic factors including gender, age and occupational background, as well as individual technical affinity. Major concerns that hinder patients' willingness to adopt AI systems stem from the black-box effect, risks to privacy and data security, insufficient emotional support, inadequate clinical validation and ambiguous regulatory accountability. Moving forward, AI application models tailored to assisted reproductive settings should be developed on the premise of rigorous clinical verification, improved ethical supervision, guaranteed data security and retained humanistic care, so as to facilitate the safe, standardized and effective integration of AI into routine clinical practice.

Key words: Artificial intelligence, Reproductive techniques, assisted, Clinical decision-making, Therapy, Patients willingness