人工智能在辅助生殖控制性卵巢刺激应用的研究进展  

Research progress on the application of artificial intelligence in controlled ovarian stimulation

作　　者：朱洁茹[1] 欧建平[1] Zhu Jieru;Ou Jianping(Center for Reproductive Medicine,Third Affiliated Hospital of Sun Yat-sen University,Guangzhou510630,China)

机构地区：[1]中山大学附属第三医院生殖医学中心,广州510630

出　　处：《中华生殖与避孕杂志》2025年第1期14-18,共5页Chinese Journal of Reproduction and Contraception

摘　　要：近年来,人工智能(artificial intelligence,AI)技术在医疗健康领域取得了广泛应用,尤其在疾病诊断与治疗决策方面带来了颠覆性变革。辅助生殖技术(assisted reproductive technology,ART)作为治疗不孕不育的重要方法,亦受益于AI技术的融入,特别是其核心环节——控制性卵巢刺激(controlled ovarian stimulation,COS)的智能化发展。传统COS方案的选择与调整高度依赖医生的经验与主观判断,存在不确定性;而AI技术通过深度学习患者的人口统计学特征、生殖内分泌水平及超声监测结果等多维度数据,为COS提供了精准的个性化优化与动态调整方案。具体而言,AI模型能够精确计算COS起始剂量、智能监测卵泡发育过程、实时预测最佳排卵触发时机,从而显著提升诊疗效率,减轻医生工作负担,并为患者提供更为个体化、精准化的治疗方案。本文对AI在COS中促性腺激素起始剂量个体化优化、卵泡发育智能监测、卵巢反应性评估及最佳排卵触发时机预测四个方面的最新研究进展进行综述,旨在为AI在辅助生殖超促排卵中的临床实践提供有价值的参考。In recent years,artificial intelligence(AI)technology has seen widespread application in the field of healthcare,particularly revolutionizing disease diagnosis and treatment decisions.Assisted reproductive technology(ART),a crucial method for treating infertility,has also benefited from the integration of AI,especially in the intelligent development of its core process--controlled ovarian stimulation(COS).Traditional COS protocols heavily relied on the experience and subjective judgment of physicians,leading to uncertainties.However,AI technology leverages deep learning to analyze multi-dimensional data,including patients'demographic characteristics,reproductive endocrine levels,and ultrasound monitoring results,to provide precise,personalized optimization and dynamic adjustments for COS.Specifically,AI models can accurately calculate the initial COS dosage,intelligently monitor follicular development,and predict the optimal timing for ovulation triggering in real-time,significantly enhancing diagnostic and treatment efficiency,reducing the workload of physicians,and offering more individualized and precise treatment plans for patients.This article reviews the latest research progress in AI applications for individualized optimization of initial gonadotropin dosage during COS,intelligent follicular monitoring,assessment of ovarian responsiveness,and prediction of the optimal timing for ovulation triggering,aiming to provide valuable insights for the clinical practice of AI in assisted reproductive hyperstimulation.

关 键 词：人工智能 机器学习 生殖技术 辅助 控制性卵巢刺激 不孕 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] R714.8[自动化与计算机技术—控制科学与工程]