Deep Metric Learning for Histopathological Image Classification

Neural networks demonstrated to be effective in multiple classification tasks with performances that are similar to human capabilities. Notwithstanding, the viability of the application of this kind of tool in real cases passes through the possibility to interpret the provided results and let the human operator take his decision according to the information that is provided. This aspect is much more evident when the field of application is bound to people's health as for biomedical image classification. We propose for the classification of histopathological images a convolutional neural network that, through metric learning, learns a representation that gathers in homogeneous clusters the labeled samples according to their characteristics. This representation, beyond improving the classification performance, also provides for the new test image sets of previously labeled samples that can be inspected to support the labeling decision. The technique has been tested on the LC25000 dataset that collects lung and colon histopathological images and with the Epistroma dataset. The source code is available at https://github.com/Calder10/Epistroma_LC25000-Classification.