Many news have been published since 2 or 3 years about Artificial Intelligence (AI) and Machine Learning / Deep Learning use in picture analysis. Lastly, Microsoft published an astonishing article where they explained that this technology is so powerful that the government needs to regulate it, especially for face recognition. In HCS Pharma we are deeply involved in picture analysis and we have worked on these technologies with the help of our partners (TISBio Lille, Molecular Devices, agence dad) since 2017.
Why is it important for us?
As we are working in 3D, we need to master all the processes in 3D: 3D biological models, 3D culture with BIOMIMESYS® products, 3D pictures acquisition with ImageXpress micro confocal systems from Molecular Devices and finally, 3D pictures analysis with MetaXpress software (also from Molecular Devices).
We get nice results with MetaXpress 3D modules…
For example, you can see below a big spheroid with well contrasted cells. In this case, it’s really easy to segment cells inside, up to 50 to 60 um.
Then, we can extract relevant parameters like cell positions, intensities of fluorescence, morphological elements, co-localization, etc.
For example, the spheroid is rebuilded on the picture (on the right) with x,y positions of each cell, z distance from the plate bottom as color and size of cells representing DAPI intensity.
… but we search for a better segmentation in complex pictures
When you mix many shapes in a 3D pictures, like cells, neurites, spots or holes, it becomes more difficult to segment precisely. Indeed, because objects are occulted by other objects, the blur in Z direction is increasing (translucient medium) and, of course, precision in the Z direction are often ten times worse than in the x/y directions.
This is why we have tried to segment our pictures with the help of WEKA (and the fiji plugin). This pixel-based segmentation shows us very interesting results compared to morphological segmentation.
For example, look at this picture of Luhmes cells in BIOMIMESYS® Brain (more explanation here) with DAPI and FITC channels. We used WEKA to separate “area of objects”: cell nuclei in DAPI and neurites in FITC. With a short training where we label areas, WEKA classified each pixel and produced a segmentation map. Results for the whole stack (from 0 to 70 microns) is visible below:
On the left picture, it is possible to isolate 4 areas:
- Blue : mostly cell nuclei structures
- Green : mostly neurites structures
- Clear Blue : both structures
- Black : nothing
Results are really promising because WEKA also gave us “probability maps” for each structure. Theses maps could be used to filter raw pictures and help to create efficient 3D reconstructions.
To masterise machine learning in picture analysis is not easy, even if some softwares seems like “push the button and look at the results” 🙂 In fact, you must investigate many variables, many methods and always get back to biological facts. But indeed, improving segmentations is a good investment!
We are now working on teaching the software how o efficiently recognize cellular and subcellular structures in complex 3D cultures.
6 Comments
Why do we use A.I and machine learning in HCS Pharma? | Skinobs · February 25, 2019 at 5:54 pm
[…] Read more… […]
Meet us at "Rendez-vous CARNOT" in Paris | HCS Pharma · October 10, 2019 at 2:16 pm
[…] present our activities in High Content Screening (HCS) and our progress in image analysis using machine learning and artificial intelligence […]
HCS Pharma is active in Asia this month! | HCS Pharma · October 16, 2019 at 2:00 pm
[…] On October 17, we will present a keynote at the 8th annual Molecular Devices user meeting in Taiwan. On October 25, we will be at the 7th Molecular Devices user meeting in Beijing. In these meetings, we will talk about challenges in 3D cell cultures and our best practices to overcome them! In particular, we will focus on the huge influence of the microenvironnement on the cells, (as mechanical stress on cancer cells) and on the response of the drugs that we observe by mofiying the composition and stiffness of the ECM thanks to BIOMIMESYS® hydroscaffold technology. We will also explain our uses of Molecular Devices tools, and especially our ImageXpress micro confocal high-content imaging systems and MetaXpress softwares. We will be happy to share also our view and our practices about pictures analysing and parameters extraction, especially in 3D context! Yes, we should talk about A.I. 🙂 […]
Artifical Intelligence and HCS in 3D cell culture: the perfect team to revolutionize drug discovery! | HCS Pharma · November 7, 2019 at 1:08 pm
[…] HCS Pharma, we have already talked about Artificial Intelligence (AI) and the way it is useful in picture analysis. AI can obviously be used for many endeavours, like the shortening of drug discovery timeline as it […]
We were at the #WestDataFestival to talk about AI! | HCS Pharma · February 11, 2020 at 2:10 pm
[…] 2017), and how a small company can do it in practice. The first day of the conference was dedicated to testing, to comparing, to learning the basics of the field. We have met many IA companies (more than 120!) to understand what can be done to increase the […]
Questions about AI/ML/DL? Look at this impressive set of articles from The Pistoia Alliance | HCS Pharma · December 8, 2020 at 4:06 pm
[…] HCS Pharma, since we have been working with machine learning to improve our pictures’ segmentation in 3D for 2 years, we now measure the difficulties to […]
Artificial Intelligence and Drug Discovery, a true love story? | HCS Pharma · December 18, 2020 at 10:49 am
[…] HCS Pharma, we use AI to improve our pictures analyzing capacities. Our first results with machine learning are very promising and we are associated to a doctoral […]