Figure 1 | Scientific Reports

Figure 1

From: Prostate cancer histopathology using label-free multispectral deep-UV microscopy quantifies phenotypes of tumor aggressiveness and enables multiple diagnostic virtual stains

Figure 1The alternative text for this image may have been generated using AI.

Processing of multi-spectral deep UV images using the two proposed methods. Top: geometrical representation of principal component analysis. Bottom: Unsupervised content-preserving transformation for optical microscopy. (a) A representative multi-spectral deep UV transmission data cube taken at 220, 255, 280 and 300 nm. (b) The 4 principal components resulting from 130 million spectra from representative select regions. (c) Calculated projections of the data cube on the principal components. (d) Schematic of data conversion from Cartesian to Spherical coordinates. (e) Representative 2D histogram of a data cube using azimuthal and elevation coordinates. (f) Schematic of trained Neural network. (g) Representative virtually H&E stained prostate tissue (Network output).

Back to article page