Blind stain separation using model-aware generative learning and its applications on fluorescence microscopy images

TL;DR

This paper introduces a novel deep learning framework for blind stain separation in biomedical images, integrating physical imaging models with generative adversarial networks to improve co-localization quantification.

Contribution

It proposes a model-aware generative learning approach that incorporates physical imaging models and a new training algorithm for fluorescence microscopy image analysis.

Findings

Outperforms prior model-based stain separation methods.

Achieves superior results over existing learning-based approaches.

Effective in fluorescence microscopy unmixing tasks.

Abstract

Multiple stains are usually used to highlight biological substances in biomedical image analysis. To decompose multiple stains for co-localization quantification, blind source separation is usually performed. Prior model-based stain separation methods usually rely on stains' spatial distributions over an image and may fail to solve the co-localization problem. With the advantage of machine learning, deep generative models are used for this purpose. Since prior knowledge of imaging models is ignored in purely data-driven solutions, these methods may be sub-optimal. In this study, a novel learning-based blind source separation framework is proposed, where the physical model of biomedical imaging is incorporated to regularize the learning process. The introduced model-relevant adversarial loss couples all generators in the framework and limits the capacities of the generative models. Further more, a training algorithm is innovated for the proposed framework to avoid inter-generator confusion during learning. This paper particularly takes fluorescence unmixing in fluorescence microscopy images as an application example of the proposed framework. Qualitative and quantitative experimentation on a public fluorescence microscopy image set demonstrates the superiority of the proposed method over both prior model-based approaches and learning-based methods.