Label-free AI-enabled Imaging Flow Cytometry for Discovery of Subpopulations of Extravillous Trophoblast from a Stem Cell-based Model of Trophoblast Differentiation

Hi, good afternoon, everyone, and also thank Louise for the invitation for me to have the opportunity to talk to this audience.

You can clearly see from the title that this is a close collaboration between Madison and engineering.

We set the record of the longest title, and probably not very comprehensible, so I will don't bother you with this title and explain what it means.

Also, I try my very best to avoid all the engineering jargons, but if I still couldn't get rid of all of them, I hope that you will forgive me in going through all this jargon, but what I try to do is to explain to you what I actually try to do.

is we actually try to develop a new and enabling biomedical devices and systems to accelerate the biomedical research, discovery, and therapeutics development.

That's what we try to do.

And we look at actually the basic unit for our organism, that's cells.

And we try to actually develop a cell analysis tools to understand and isolate the cells for a deeper analysis of the cells.

And the far-fetched dream is that can we actually look at the cells, and understand its genome, understand its proteomics, understand its metabolics.

These are very far-fetched dreams, but we hope that we could develop some tools that give you a chance to look into that direction, aided by the very fast development of AI technology.

So that's what we try to do.

The first few slides we're going to actually, you know, go through a very quick introduction about why we come up with these tools.

What you see here on the left is a microscope.

On the right is a flow cytometer or FAC system fluorescence activated cell sorting.

Microscope gives us a beautiful image of every single cell.

It has all the information, morphological features, phenotypical features of the cells.

But it lacks one function that is very desirable and important.

That is, can we isolate the cell of particular interest and take it out from the cell? the sample and do the downstream molecular analysis.

It's very difficult to do it.

We have to use microlaser dissection.

That is a very slow process and damages the cells most of the time.

You can handle maybe a few cells in research lab, but certainly not suitable for clinicals.

On the right, you see a fluorescent activated cell sorter.

The good thing is you suspend the cells in the fluid and you flow the cells in one single fire, in one single land.

You use high-speed laser to interrogate every single of them.