Fig. 2: Convergence of the overall configuration in UMAP space as the number of ROIs increases. | Communications Engineering

Fig. 2: Convergence of the overall configuration in UMAP space as the number of ROIs increases.

From: Unsupervised learning of pixel clustering in Mueller matrix images for mapping microstructural features in pathological tissues

Fig. 2

The numbers indicate the fraction of data used, increasing at a step of 1/6, equivalent to 37 regions of interest (ROI) from samples of 6–7 patients. Above the numbers are the corresponding Uniform Manifold Approximation and Projection (UMAP) using the specified fraction of total data.

Back to article page