Virtual tissue staining by deep learning

Our goal is to create a virtual tissue-staining device to replace manual chemical-staining techniques by integrating optics and deep

learning.

Chemical staining of cell components plays an essential role in biomedical and pharmaceutical research and practice. Cell structures

of interest are highlighted using various chemical stains and imaged with the appropriate optical setup. However, these techniques

are often invasive and sometimes even toxic to the cells, in addition to being time consuming, labor intensive, and expensive.

Recently, the use of deep learning has been proposed as a way to create images of virtually stained cell structures, thus mitigating the

inherent problems associated with the conventional chemical staining. However, these methods are usually specialized for a specific

application and, thus, highly dependent on the setting of the optical device used for acquiring the training data.

In order to make virtual tissue-staining more accessible to end-users, we propose a device based on a simple optical system with an

integrated deep-learning-powered virtual-staining software.

This is an ideal moment to enter the virtual tissue-staining market because we can gain a first mover advantage. Further, we can take

advantage of the fact that the tissue-staining market is expected to grow with a compound annual rate of roughly 8.5% until 2025 (up

to 3400M USD), and to continue to grow for the foreseeable future.

As part of this project, we aim to launch the startup company IFLAI to commercially exploit our virtual-staining technology and the

prototype we will develop. With the startup IFLAI, we aim to provide ~20 jobs to university-educated individuals in the EU within the

next 5 years. IFLAI has already received initial funding and support from two different organisations that support and believe in its

venture.