Comparison of time series clustering methods for identifying novel subphenotypes of patients with infection

Sivasubramanium V Bhavani; Li Xiong; Abish Pius; Matthew Semler; Edward T Qian; Philip A Verhoef; Chad Robichaux; Craig M Coopersmith; Matthew M Churpek

doi:10.1093/jamia/ocad063

Comparison of time series clustering methods for identifying novel subphenotypes of patients with infection

J Am Med Inform Assoc. 2023 May 19;30(6):1158-1166. doi: 10.1093/jamia/ocad063.

Authors

Sivasubramanium V Bhavani^{1

2}, Li Xiong³, Abish Pius⁴, Matthew Semler⁵, Edward T Qian⁵, Philip A Verhoef^{6

7}, Chad Robichaux⁸, Craig M Coopersmith^{2

9}, Matthew M Churpek^{10

11}

Affiliations

¹ Department of Medicine, Emory University, Atlanta, Georgia, USA.
² Emory Critical Care Center, Atlanta, Georgia, USA.
³ Department of Computer Science, Emory University, Atlanta, Georgia, USA.
⁴ Department of Computational & Systems Biology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA.
⁵ Department of Medicine, Vanderbilt University, Nashville, Tennessee, USA.
⁶ Department of Medicine, University of Hawaii John A. Burns School of Medicine, Honolulu, Hawaii, USA.
⁷ Hawaii Permanente Medical Group, Honolulu, Hawaii, USA.
⁸ Department of Biomedical Informatics, Emory University, Atlanta, Georgia, USA.
⁹ Department of Surgery, Emory University, Atlanta, Georgia, USA.
¹⁰ Department of Medicine, University of Wisconsin, Madison, Wisconsin, USA.
¹¹ Department of Biostatistics and Medical Informatics, University of Wisconsin, Madison, Wisconsin, USA.

Abstract

Objective: Severe 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 methods: Patients 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.

Results: There 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.

Discussion: DTW- and GBTM-based clustering algorithms applied to vital signs in infected patients identified consistent subphenotypes with distinct clinical outcomes and treatment responses.

Conclusion: Time series clustering with distinct computational approaches demonstrate similar performance and significant agreement in the resulting subphenotypes.

Keywords: infection; phenotypes; sepsis; subphenotypes; vital signs.

Publication types

Research Support, U.S. Gov't, Non-P.H.S.
Research Support, N.I.H., Extramural

MeSH terms

Algorithms*
Cluster Analysis
Fever*
Humans
Patients
Time Factors

Abstract

Publication types

MeSH terms

Grants and funding