Deep CNN for single-cell classification based on pH fingerprints

The research goal was to leverage intracellular structural features together with intracellular pH patterns in order to classify different types of bladder cancer cells using Deep Learning. Specifically, the Deep Convolutional Neural Network (CNN)-based strategy was used to classify individual cells from the RT4 and J82 bladder cancer cell lines upon staining with a pH-sensitive dye (Bromothymol Blue) and bright-field imaging using a digital color camera mounted on a standard inverted microscope. J82 and RT4 cells were imaged using a 40x objective after the internalization of Bromothymol Blue.