Objective To investigate the relationship between the similarity of AIDS patients and their antiviral treatment regimens in order to provide reference to the personalized treatment for AIDS patients.Methods Clinical data of 648 AIDS patients who received highly active antiretroviral therapy with good immune reconstruction were collected.Patient similarity was calculated based on Euclidean distance.The most adopted treatment regimen among the top 10％ most similar patients was regarded as the recommended regimen.Results of case-based reasoning and cluster analysis were presented by data visualization tools.Results Treatment regimens recommended on the basis of patient similarities had up to 83.3％ agreement with the initial treatment regimens in practice.Conclusions Recommended treatment regimens based on the patient similarities were highly consistent with the actual regimens,confirming the feasibility and effectiveness of the case-based reasoning and cluster analysis in the personalized treatment for AIDS patients,which provided a way for the further precision medicine for AIDS patients.%目的 探讨艾滋病患者相似性与抗病毒治疗方案之间的关系,为制定艾滋病患者个性化治疗方案提供依据.方法 收集首次接受抗病毒治疗且治疗后免疫重建良好的648例艾滋病患者临床资料,计算基于欧氏距离的患者相似性,以最相似的前10％患者治疗方案作为推荐方案,实现基于案例推理的治疗方案推荐,并通过数据可视化工具呈现基于案例推理和聚类分析的结果.结果 基于患者相似性推荐的治疗方案与实际初始治疗方案的一致率达到83.3％.结论 根据患者相似性推荐的治疗方案与实际的治疗方案之间具有较高的一致性,证实了基于案例推理和聚类分析在艾滋病患者个性化治疗方案推荐中的可行性和有效性,为艾滋病领域的精准医疗研究提供了参考.
展开▼
