Comparison of time series clustering methods for identifying novel subphenotypes of patients with infection.
Comparison of time series clustering methods for identifying novel subphenotypes of patients with infection.
复制标题
用于识别感染患者新亚表型的时间序列聚类方法的比较。
DOI：
10.1093/jamia/ocad063
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
Churpek,MatthewM
中科院分区：
文献类型：
--
作者：
Bhavani,SivasubramaniumV;Xiong,Li;Pius,Abish;Semler,Matthew;Qian,EdwardT;Verhoef,PhilipA;Robichaux,Chad;Coopersmith,CraigM;Churpek,MatthewM
目的严重感染可导致器官功能障碍和败血症。识别感染患者的亚表型对于个性化管理至关重要。目前尚不清楚不同的时间序列聚类算法在识别这些亚表型方面有何不同。 材料和方法 2014 年至 2019 年期间入住埃默里医疗机构 4 家医院的疑似感染患者被纳入其中，分为单独的训练和验证队列。对住院前 8 小时的生命体征应用动态时间规整 (DTW)，并使用层次聚类 (DTW-HC) 和围绕中心点的分区 (DTW-PAM) 将患者聚类为亚表型。对 DTW-HC、DTW-PAM 和之前发布的基于组的轨迹模型 (GBTM) 进行了评估，以了解亚表型簇、轨迹模式以及亚表型与临床结果和治疗反应的关联。结果 训练中有 12 473 名患者，验证队列中有 8256 名患者。 DTW-HC、DTW-PAM 和 GBTM 模型产生了 4 种一致的生命体征轨迹模式，在聚类中具有显着一致性（71-80% 一致性，P< .001）：A 组为体温过高、心动过速、呼吸急促和低血压。 B组为高热、心动过速、呼吸急促和高血压。 C组和D组的体温、心率和呼吸频率较低，C组血压正常，D组血压较低。在所有 3 个模型中，与生理盐水相比，A 组的 30 天住院患者死亡率比值比较高 (P< .01)，D 组的死亡率显着优于生理盐水 (P< .01)。讨论应用于受感染患者生命体征的基于 DTW 和 GBTM 的聚类算法识别出具有不同临床结果和治疗反应的一致亚表型。结论采用不同计算方法的时间序列聚类表现出相似的性能和结果的显着一致性。亚表型。
ObjectiveSevere infection can lead to organ dysfunction and sepsis. Identifying subphenotypes of infected patients is essential for personalized management. It is unknown how different time series clustering algorithms compare in identifying these subphenotypes.Materials and MethodsPatients with suspected infection admitted between 2014 and 2019 to 4 hospitals in Emory healthcare were included, split into separate training and validation cohorts. Dynamic time warping (DTW) was applied to vital signs from the first 8 h of hospitalization, and hierarchical clustering (DTW-HC) and partition around medoids (DTW-PAM) were used to cluster patients into subphenotypes. DTW-HC, DTW-PAM, and a previously published group-based trajectory model (GBTM) were evaluated for agreement in subphenotype clusters, trajectory patterns, and subphenotype associations with clinical outcomes and treatment responses.ResultsThere were 12 473 patients in training and 8256 patients in validation cohorts. DTW-HC, DTW-PAM, and GBTM models resulted in 4 consistent vitals trajectory patterns with significant agreement in clustering (71–80% agreement,P< .001): group A was hyperthermic, tachycardic, tachypneic, and hypotensive. Group B was hyperthermic, tachycardic, tachypneic, and hypertensive. Groups C and D had lower temperatures, heart rates, and respiratory rates, with group C normotensive and group D hypotensive. Group A had higher odds ratio of 30-day inpatient mortality (P< .01) and group D had significant mortality benefit from balanced crystalloids compared to saline (P< .01) in all 3 models.DiscussionDTW- and GBTM-based clustering algorithms applied to vital signs in infected patients identified consistent subphenotypes with distinct clinical outcomes and treatment responses.ConclusionTime series clustering with distinct computational approaches demonstrate similar performance and significant agreement in the resulting subphenotypes.