Figure 3
From: Visual parameter optimisation for biomedical image processing
Interactive sorting of input parameters of a colour deconvolution algorithm applied to a stained histology image of a liver section (see case study). (a) Applying "smart sorting" identifies input parameter p2 as the one with the highest aggregate correlation with variance of the output measures and sorts the rows of data according to values assumed for p2. This yields a step-like pattern with a bin for each unique value that p2 takes. Also, correlations between p2 and the output measures emerge, for example, p2 is directly correlated with m2 and inversely correlated with m1, m3, and m6. (b) Subsequent sorting on p1 reveals even more striking patterns. For example, in addition to the direct correlation with p2, m2 is also inversely correlated with p1.