Talk 25 Jan 2019: Cédric ALLIER (CEA-Leti Grenoble) on Deep learning technologies for living cells microscopy

1 view
Skip to first unread message

Jakob Verbeek

unread,
Dec 15, 2018, 4:49:29 PM12/15/18
to smile-in...@googlegroups.com

Talk 25 Jan 2019, 11am, INRIA Grenoble (Montbonnot), F107

Deep learning technologies for living cells microscopy
Dr. Cédric ALLIER (CEA-Leti Grenoble)

In the present talk, I will discuss the recent development of deep 
learning technologies applied to microscopy techniques of living cells 
(2D and 3D cell culture, organoids, tissue etc.). A large number of 
publications appear recently discussing the advantages of CNN applied to 
microscopy, to e.g. improve the image quality, reduce the acquisition 
time, reduce the number of frame (super-resolution, 3D tomographic 
reconstruction), to perform cell segmentation and  classification, etc.
In particular I will discuss the combination of deep learning and 
lens-free microscopy, a technique that we are developing since 2009 at 
CEA-Leti. This technique relies already heavily on computation, since in 
this setup the optics are replaced by holographic reconstruction 
algorithms.  In comparison with existing techniques, this microscope 
combines several advantages: it is very small, cheap, yet it provides an 
extremely large field of view (30 mm2) showing at a glance N>20.000 
cells. Recently, we showed, for the first time, 3D+time acquisitions of 
3D cell culture with a lens-free microscope. We obtained time-lapse 
views of 3D culture of prostate epithelial cells showing complex 
cellular self-organizations that occurred through migration of single 
cells, displacement of large clusters, and merging and interconnection 
over long distances (>1 mm). Today, our team is pushing further the 
computational aspect of holographic microscopy by integrating 
deep-learning technologies. Our aim is to develop algorithms that will 
help the biologists and clinicians to perform automatic acquisitions and 
analysis of cultured cells, cultured tumors, and tissue.

Jakob Verbeek

unread,
Jan 17, 2019, 3:49:43 AM1/17/19
to smile-in...@googlegroups.com
--
Jakob Verbeek
+33.6.28.06.31.36

Jakob Verbeek

unread,
Jan 24, 2019, 9:18:59 AM1/24/19
to smile-in...@googlegroups.com
Reminder of a talk tomorrow 11am @ INRIA Montbonnot

On Sat, Dec 15, 2018 at 10:35 PM Jakob Verbeek <j.j.v...@gmail.com> wrote:


--
Jakob Verbeek
+33.6.28.06.31.36
Reply all
Reply to author
Forward
0 new messages