The application of artificial intelligence (AI) methods in medical field is increasing year by year; however, few studies have applied AI methods in the reproductive field. In view of the complexity of infertility diagnosis and treatment, a machine learning‐based risk scoring system for infertility was constructed in this paper to help clinicians better grasp the patient's condition. First, eight key features of infertility are screened out by feature selection. Second, the entropy‐based feature discretization method was used to divide the feature abnormal intervals, and the random forest was used to determine the weight of each feature. Finally, the pregnancy outcome can be predicted according to the overall risk score of patients, which is helpful for doctors to choose targeted treatment more efficiently. It is worth noting that, to further improve the accuracy of the diagnosis, we also divided the patients into age groups and constructed the corresponding risk scoring system for patients of different age groups. The stability test results show the good performance of the system. The risk scoring system for infertility built in this paper is a meaningful exploration of the application of AI in the field of reproduction.

中文翻译：

---

基于机器学习的不育症风险评分系统，考虑了不同年龄组

人工智能（AI）方法在医学领域的应用逐年增加。但是，很少有研究将AI方法应用于生殖领域。鉴于不孕症诊断和治疗的复杂性，本文构建了基于机器学习的不孕症风险评分系统，以帮助临床医生更好地掌握患者的病情。首先，通过特征选择筛选出不孕症的八个关键特征。其次，使用基于熵的特征离散化方法来划分特征异常间隔，并使用随机森林来确定每个特征的权重。最后，可以根据患者的整体风险评分来预测妊娠结局，这有助于医生更有效地选择靶向治疗。值得一提的是，为了进一步提高诊断的准确性，我们还将患者分为年龄组，并针对不同年龄组的患者建立了相应的风险评分系统。稳定性测试结果表明系统性能良好。本文构建的不孕风险评分系统是对AI在生殖领域的应用的有意义的探索。