Skip to main content
. 2021 Jun 1;10(11):2450. doi: 10.3390/jcm10112450

Figure 2.

Figure 2

Cluster analysis identified two phenogroups of former smokers. In order to identify phenogroups within the population of former smokers, Gower distance to identify dissimilarity between individuals was calculated and a partitioning around mediods algorithm was used to identify the clusters. The clusters were visualized using t-distributed stochastic neighborhood embedding to reduce dimensionality of the data.