在过去的20年里，中国的不孕率从3%上升到12.5%至15%。不孕症已成为继癌症和心血管疾病之后的第三大疾病。随着时间的推移，试管受精和胚胎移植（IVF-ET）在不孕治疗领域变得越来越重要。然而，IVF-ET的成功率报告为30%-40%，而且费用逐渐上升。同时，为了提高成功率和降低费用，选择最佳的IVF-ET治疗策略至关重要。在临床工作中，IVF-ET治疗策略的选择往往是基于医生的经验而没有统一的标准。为了解决这个重要而复杂的问题，我们提出了一种基于人工智能（AI）的最佳治疗策略选择系统，通过模拟IVF-ET过程并分析大量的治疗决策，从临床数据中提取隐含的知识，用于新的和复诊患者。我们证明了该模型在10个AI分类算法中的性能存在差异。因此，我们需要选择最佳的方法来预测不同IVF-ET治疗策略下患者的妊娠结果。此外，通过提出的模型确定了特征排名，以衡量每个患者特征的重要性。从而可以为个体患者特征提供更好的建议，医生可以针对特定患者特征提供更有效的建议，以提高诊断的准确性和效率。©2023. 作者在Springer Science+Business Media, LLC,部分Springer Nature独家许可下发表此论文。

Over the last 20 years, China's infertility rate has risen from 3% to 12.5%-15%. Infertility has become the third largest disease following cancer and cardiovascular disease. Then, the in vitro fertilization and embryo transfer (IVF-ET) becomes more and more important in infertility treatment field. However, the reported success rate for IVT-ET is 30%-40% and costs are gradually rising. Meanwhile, to increase success rates and decrease costs, the optimal selection of the IVF-ET treatment strategy is crucial. In a clinical work, the IVF-ET treatment strategy selection is always based on the experience of the doctor without a uniform standard. To solve this important and complex problem, we proposed an artificial intelligence (AI)-based optimal treatment strategy selection system to extract implicit knowledge from clinical data for new and returning patients, by mimicking the IVF-ET process and analysing a myriad of treatment decisions. We demonstrated that the performance of the model was different in 10 AI classification algorithms. Hence, we need to select the optimal method for predicting patient pregnancy result in different IVF-ET treatment strategies. Moreover, feature ranking is determined in the proposed model to measure the importance of each patient characteristics. Therefore, better advice can be provided for individual patient characteristics, doctors can provide more valid suggestions regarding certain patient characteristics to improve the accuracy of diagnosis and efficiency.© 2023. The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.