Limits...
Identification of complex metabolic states in critically injured patients using bioinformatic cluster analysis.

Cohen MJ, Grossman AD, Morabito D, Knudson MM, Butte AJ, Manley GT - Crit Care (2010)

Bottom Line: While patients transitioned between states, they spent significant amounts of time in each.Further analysis of correlations between pairs of variables within each cluster reveals significant differences in physiology between clusters.These results demonstrate that hierarchical clustering techniques can be useful for visualizing complex multivariate data and may provide new insights for the care of critically injured patients.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Surgery, University of California, 505 Parnassus Avenue, San Francisco, CA 94143, USA. mcohen@sfghsurg.ucsf.edu

ABSTRACT

Introduction: Advances in technology have made extensive monitoring of patient physiology the standard of care in intensive care units (ICUs). While many systems exist to compile these data, there has been no systematic multivariate analysis and categorization across patient physiological data. The sheer volume and complexity of these data make pattern recognition or identification of patient state difficult. Hierarchical cluster analysis allows visualization of high dimensional data and enables pattern recognition and identification of physiologic patient states. We hypothesized that processing of multivariate data using hierarchical clustering techniques would allow identification of otherwise hidden patient physiologic patterns that would be predictive of outcome.

Methods: Multivariate physiologic and ventilator data were collected continuously using a multimodal bioinformatics system in the surgical ICU at San Francisco General Hospital. These data were incorporated with non-continuous data and stored on a server in the ICU. A hierarchical clustering algorithm grouped each minute of data into 1 of 10 clusters. Clusters were correlated with outcome measures including incidence of infection, multiple organ failure (MOF), and mortality.

Results: We identified 10 clusters, which we defined as distinct patient states. While patients transitioned between states, they spent significant amounts of time in each. Clusters were enriched for our outcome measures: 2 of the 10 states were enriched for infection, 6 of 10 were enriched for MOF, and 3 of 10 were enriched for death. Further analysis of correlations between pairs of variables within each cluster reveals significant differences in physiology between clusters.

Conclusions: Here we show for the first time the feasibility of clustering physiological measurements to identify clinically relevant patient states after trauma. These results demonstrate that hierarchical clustering techniques can be useful for visualizing complex multivariate data and may provide new insights for the care of critically injured patients.

Show MeSH

Related in: MedlinePlus

Correlations of pairs of variables between clusters 1 (live) and 4 (die). Cluster 1 is shown in blue and cluster 4 in green. Correlation coefficients are shown on the lines and the variables above each plot.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2875524&req=5

Figure 6: Correlations of pairs of variables between clusters 1 (live) and 4 (die). Cluster 1 is shown in blue and cluster 4 in green. Correlation coefficients are shown on the lines and the variables above each plot.

Mentions: Having determined that 1) univariate analysis did not provide adequate predictors and 2) that hierarchical clustering provided superior prediction of outcomes, we next sought to determine why this was the case. We hypothesized that the clusters contained new physiological relationships and that the correlations between variable pairs would differ according to patient state. Furthermore, we believed that these changing correlations would likely reflect changing physiological relationships depending on the changing injury or resuscitation state of a patient. To test this we next examined the correlations between pairs of variables within each cluster. To confirm that our correlations were statistically significant, we performed bootstrap resampling and label shuffling. Figure 6 shows the correlation coefficients of variable pairs for cluster 4, the cluster most closely associated with death, and cluster 1, which was most closely associated with good outcome. Examination of these results was very revealing and provided proof of both the discrimination of the clustering technique and the ability of this technique to identify physiologic relationships that would otherwise be impossible to discern.


Identification of complex metabolic states in critically injured patients using bioinformatic cluster analysis.

Cohen MJ, Grossman AD, Morabito D, Knudson MM, Butte AJ, Manley GT - Crit Care (2010)

Correlations of pairs of variables between clusters 1 (live) and 4 (die). Cluster 1 is shown in blue and cluster 4 in green. Correlation coefficients are shown on the lines and the variables above each plot.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2875524&req=5

Figure 6: Correlations of pairs of variables between clusters 1 (live) and 4 (die). Cluster 1 is shown in blue and cluster 4 in green. Correlation coefficients are shown on the lines and the variables above each plot.
Mentions: Having determined that 1) univariate analysis did not provide adequate predictors and 2) that hierarchical clustering provided superior prediction of outcomes, we next sought to determine why this was the case. We hypothesized that the clusters contained new physiological relationships and that the correlations between variable pairs would differ according to patient state. Furthermore, we believed that these changing correlations would likely reflect changing physiological relationships depending on the changing injury or resuscitation state of a patient. To test this we next examined the correlations between pairs of variables within each cluster. To confirm that our correlations were statistically significant, we performed bootstrap resampling and label shuffling. Figure 6 shows the correlation coefficients of variable pairs for cluster 4, the cluster most closely associated with death, and cluster 1, which was most closely associated with good outcome. Examination of these results was very revealing and provided proof of both the discrimination of the clustering technique and the ability of this technique to identify physiologic relationships that would otherwise be impossible to discern.

Bottom Line: While patients transitioned between states, they spent significant amounts of time in each.Further analysis of correlations between pairs of variables within each cluster reveals significant differences in physiology between clusters.These results demonstrate that hierarchical clustering techniques can be useful for visualizing complex multivariate data and may provide new insights for the care of critically injured patients.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Surgery, University of California, 505 Parnassus Avenue, San Francisco, CA 94143, USA. mcohen@sfghsurg.ucsf.edu

ABSTRACT

Introduction: Advances in technology have made extensive monitoring of patient physiology the standard of care in intensive care units (ICUs). While many systems exist to compile these data, there has been no systematic multivariate analysis and categorization across patient physiological data. The sheer volume and complexity of these data make pattern recognition or identification of patient state difficult. Hierarchical cluster analysis allows visualization of high dimensional data and enables pattern recognition and identification of physiologic patient states. We hypothesized that processing of multivariate data using hierarchical clustering techniques would allow identification of otherwise hidden patient physiologic patterns that would be predictive of outcome.

Methods: Multivariate physiologic and ventilator data were collected continuously using a multimodal bioinformatics system in the surgical ICU at San Francisco General Hospital. These data were incorporated with non-continuous data and stored on a server in the ICU. A hierarchical clustering algorithm grouped each minute of data into 1 of 10 clusters. Clusters were correlated with outcome measures including incidence of infection, multiple organ failure (MOF), and mortality.

Results: We identified 10 clusters, which we defined as distinct patient states. While patients transitioned between states, they spent significant amounts of time in each. Clusters were enriched for our outcome measures: 2 of the 10 states were enriched for infection, 6 of 10 were enriched for MOF, and 3 of 10 were enriched for death. Further analysis of correlations between pairs of variables within each cluster reveals significant differences in physiology between clusters.

Conclusions: Here we show for the first time the feasibility of clustering physiological measurements to identify clinically relevant patient states after trauma. These results demonstrate that hierarchical clustering techniques can be useful for visualizing complex multivariate data and may provide new insights for the care of critically injured patients.

Show MeSH
Related in: MedlinePlus