Identification of Sepsis Subphenotypes via ICU Data Clustering

• Giovana Assis UFMG
• Victoria F. Mello UFMG
• Haniel B. Ribeiro UFMG
• Alexandre G. Barros UFMG
• Gisele L. Pappa UFMG
• Wagner Meira Jr. UFMG

DOI: https://doi.org/10.5753/eniac.2024.245281

Abstract

This study aims to identify sepsis subphenotypes in ICU patients using clustering techniques on MIMIC-IV clinical data. Missing data were imputed using MissForest. We applied UMAP (trustworthiness = 0.97) for dimensionality reduction and K-means (K = 5, silhouette score = 0.30) for clustering, revealing five distinct subphenotypes with unique clinical characteristics. Our findings suggest that these subphenotypes could guide efficient diagnosis and personalized treatments, surpassing the current SOFA score’s limitations. The results highlight the need for further exploration of additional variables to improve sepsis diagnosis and treatment.

Keywords: Clustering, Sepsis, UTI

References

Fohner, A. E., Greene, J. D., Lawson, B. L., Chen, J. H., Kipnis, P., Escobar, G. J., and Liu, V. X. (2019). Assessing clinical heterogeneity in sepsis through treatment patterns and machine learning. Journal of the American Medical Informatics Association, 26(12):1466–1477.

Hu, C., Li, Y., Wang, F., et al. (2022). Application of machine learning for clinical subphenotype identification in sepsis. Infectious Disease Therapy, 11:1949–1964.

Ibrahim, Z. M., Wu, H., Hamoud, A., Stappen, L., Dobson, R. J. B., and Agarossi, A. (2020). On classifying sepsis heterogeneity in the icu: Insight using machine learning. Journal of the American Medical Informatics Association, 27(3):437–443.

Johnson, A. E. W., Bulgarelli, L., Shen, L., Gayles, A., Shammout, A., Horng, S., Pollard, T. J., Hao, S., Moody, B., Gow, B., Lehman, L. H., Celi, L. A., and Mark, R. G. (2023). Mimic-iv, a freely accessible electronic health record dataset. Scientific Data, 10(1):1.

Koutroulis, I., Velez, T., Wang, T., Yohannes, S., Galarraga, J. E., Morales, J. A., Freishtat, R. J., and Chamberlain, J. M. (2022). Pediatric sepsis phenotypes for enhanced therapeutics: An application of clustering to electronic health records. Journal of the American College of Emergency Physicians Open, 3(1):e12660.

McInnes, L., Healy, J., and Melville, J. (2018). Umap: Uniform manifold approximation and projection. Journal of Open Source Software, 3(29):861.

Sakr, Y., Jaschinski, U., Wittebole, X., Szakmany, T., Lipman, J., Ñamendys Silva, S. A., Martin-Loeches, I., Leone, M., Lupu, M.-N., Vincent, J.-L., and Investigators, I. (2018). Sepsis in intensive care unit patients: Worldwide data from the intensive care over nations audit. Open Forum Infectious Diseases, 5(12).

Seymour, C. W., Kennedy, J. N., Wang, S., et al. (2019). Derivation, validation, and potential treatment implications of novel clinical phenotypes for sepsis. JAMA, 321(20):2003–2017.

Singer, M., Deutschman, C., Seymour, C., Shankar-Hari, M., Annane, D., Bauer, M., Bellomo, R., Bernard, G., Chiche, J., Coopersmith, C., Hotchkiss, R., Levy, M., Marshall, J., Martin, G., Opal, S., Rubenfeld, G., van der Poll, T., Vincent, J., and Angus, D. (2016). The third international consensus definitions for sepsis and septic shock (sepsis-3). JAMA, 315(8):801–810.