Abstract
背景囊性纤维化(CF)是一种多系统疾病,其单独评估基于肺功能的疾病严重程度可能不合适。该研究的目的是开发一种全面的机器学习算法,以评估患者儿童肺功能的临床状态。
Methods综合预期收集的临床数据库(加拿大多伦多)用于应用无监督的聚类分析。然后通过当前和未来的肺功能,未来住院风险以及用口服抗生素治疗的未来肺癌(PEX)的风险来比较了定义的簇。k最近邻居(knn)算法用于潜在分配群集。这些方法在伟大的奥蒙德街医院(GOSH)的儿科临床CF数据集中验证。
ResultsThe optimal cluster model identified four (A-D) phenotypic clusters based on 12 200 encounters from 530 individuals. Two clusters (A,B) consistent with mild disease were identified with high FEV1那and low risk of both hospitalisation and PEx treated with oral antibiotics. Two clusters (C,D) consistent with severe disease were also identified with low FEV1。Cluster D对住院治疗和用口服抗生素治疗的患者和PEX进行了最短的时间。结果是在191名191名儿童遇到的191名儿童遇到的结果一致。KNN集群分配错误率低,2.5%(多伦多)和3.5%(GOSH)。
Conclusion机器学习衍生的表型簇可以预测肺功能独立的疾病严重程度,并且可以与功能措施结合使用,以预测CF患者的未来疾病术。
Footnotes
This manuscript has recently been accepted for publication in the欧洲呼吸杂志。它在汇款和排版的抄本之前在此处发布于其已接受的表单。在这些生产过程完成后,作者已批准所产生的证据,这篇文章将转向最新问题收获online. Please open or download the PDF to view this article.
Conflict of interest: Ms. Filipow has nothing to disclose.
Conflict of interest: Dr. Davies reports personal fees from Chiesi Limited, outside the submitted work.
Conflict of interest: Professor Main has nothing to disclose.
Conflict of interest: Dr. sebire has nothing to disclose.
Conflict of interest: Dr. Wallis has nothing to disclose.
利益冲突:博士鼠jen reports grants and personal fees from Vertex, Calithera, Proteostasis, TranslateBio, Genentech, Bayer and Boehringer Ingelheim, outside the submitted work.
Conflict of interest: Dr. Stanojevic reports grants from SickKids Foundation, grants from European Respiratory Society, during the conduct of the study.
- Received2020年7月23日。
- 公认December 22, 2020.
- ©作者2021.用于再生权和权限联系权限{exernet.org