Fig. 6. Hierarchical cluster analysis (HCA) for 19 TICs at IDLH concentrations and a control.

HCA is a routine model-free statistical classification method based on Euclidean distance³⁷–⁴⁰. In these experiments, the Euclidean distances are defined by the color difference vectors of each experimental trial in the full 108-dimensional space made up of the changes in red, green, and blue values of the 36 nanoporous pigments in the sensor array. As shown, all experiments were run in septuplicate: no confusions or errors in classification were observed in 140 trials. The IDLH concentrations of each analyte are shown in ppm. The HCA used minimum variance (i.e., “Ward’s Method”) for clustering.