yu fu Profile
yu fu

@yufu0413

Followers
140
Following
106
Media
0
Statuses
20

Data scientist, cancer research, machine learning

Joined July 2016
Don't wanna be here? Send us removal request.
@yufu0413
yu fu
3 years
Exciting time for better treating BC patients!
@OwkinScience
Owkin
3 years
In our blog, Senior Data Scientist @yufu0413 explains how new drugs and #AI are revolutionizing HER2-low #breastcancer therapy, including our collaboration with @GSTTresearch and @KingsCollegeLon to develop #AI-driven solutions to detect HER2-low tumors. https://t.co/JZAxKVs6K4
0
1
5
@MoritzGerstung
Moritz Gerstung
4 years
My new lab @DKFZ is recruiting! 2 experimental postdocs in spatial genomics and single cell mutagenesis + 1 research technician post to fill. Join us in beautiful Heidelberg. Please RT https://t.co/u73gu2a5RZ https://t.co/1k0CMwgLTG
0
62
91
@yufu0413
yu fu
5 years
See tumor evolution through spatial genomics+transcriptomics, magic indeed! Inspiring work by @LomakinAI @MoritzGerstung @LucyYat47076319 and colleagues.
@MoritzGerstung
Moritz Gerstung
5 years
Want to *see* how a tumour has evolved and grown? And also how different clones acquired characteristic transcriptional and histopathological features? Hold on, that's magic. No, it's our new preprint by @LucyYat47076319 and @MatsNilssonLab 1/9 https://t.co/28mjn7sAhN
0
1
3
@yufu0413
yu fu
5 years
Want to do cool science? This is the next place to be! If you are passionate about cancer research and excited to apply/develop new cutting edge tools, please join!
@MoritzGerstung
Moritz Gerstung
5 years
This summer we'll be opening a new lab at the German Cancer Research Centre @DKFZ in Heidelberg. Come join us studying cancer evolution with spatial & single cell genomics + AI. There will be plenty of openings for students, postdocs, and technicians; dry and wet lab. >>
0
0
7
@yufu0413
yu fu
5 years
With an AI assist tool, 7% increase in agreement between pathologists in grading Gleason grade-1 and 13% time saved per biopsie. Very promising
0
0
2
@yufu0413
yu fu
5 years
Congrats @SarahKillcoyne and team! Great read, thanks!
0
0
0
@yufu0413
yu fu
5 years
I have learned and grown so much during my time at @emblebi. Big thank you to @MoritzGerstung and all the lab members @NadezdaVolkova1 @zga_aaa @alex_w_jung @harald_voeh, Sarah, Rui, Santi, Stefan. Will miss you all very much!
@MoritzGerstung
Moritz Gerstung
5 years
@yufu0413 @oWkin All the best for your future journey! We will miss you.
0
0
4
@yufu0413
yu fu
5 years
Had a lot of fun during the lock down 😊 thanks to this great idea from @MoritzGerstung
@MoritzGerstung
Moritz Gerstung
5 years
@jnkath @NatureCancer Made with original data! Credit to @yufu0413 who helped with the mosaic and Spencer Philips who drew the DNA.
0
1
4
@yufu0413
yu fu
5 years
Excited to be joining a great team @OWKIN and continuing to contribute to the progress of digital pathology+cancer genomics
@OwkinScience
Owkin
5 years
Owkin is very active in the field of biomedical AI: we're thrilled to welcome @yufu0413 in our Translational Research team. She is the primary author of a paper recently published in @NatureCancer Cancer about pan-cancer computational histopathology. https://t.co/gUpR9an0Dk 3/4
1
0
12
@yufu0413
yu fu
6 years
Thanks!
@obahcall
Orli Bahcall
6 years
#BoG20 Nice talk Yu Fu (with @MoritzGerstung) on Pan-cancer computational histology (PC-CHiP) to capture tissue specific morphology, classify origin, composition. Features correlate w recurrent genetic alterations, including WGD. Use to improve prediction clinical outcomes.
0
0
2
@MoritzGerstung
Moritz Gerstung
6 years
Attending #BoG20 and interested in cancer? You shouldn't miss @yufu0413 talking later today about how she fused digital pathology and cancer genomics - "Pan-cancer computational histopathology reveals mutations, tumor composition and prognosis" https://t.co/jEWIcsQokv 1/3
1
2
35
@yufu0413
yu fu
6 years
Great work! Congrats @luiza_moore
@Luiza_Moore
Luiza Moore
6 years
Absolutely thrilled that our work on somatic mutations in normal endometrium is out in @Nature today! Using laser-capture microscopy and low input whole-genome sequencing we explored genomic and evolutionary landscapes of this fascinating tissue. [1/10] https://t.co/LARU0QPN4B
0
0
1
@yufu0413
yu fu
6 years
Thank you! @FAndreMD
@FAndreMD
FabriceAndre
6 years
congrat @yufu0413 for this outstanding paper. could digital hes slides contain everything we need to know to make decisions ?
0
0
1
@yufu0413
yu fu
6 years
Such an honor and so much more to be done!
@EricTopol
Eric Topol
6 years
Taking deep learning #AI of #cancer pathology to a new, unprecedented level: accurately classify 28 types, genomic alterations, and prognosis https://t.co/s9mP3yFr7B by @MoritzGerstung @yufu0413 @luiza_moore @emblebi and collaborators <-congrats for leading the field forward!
0
2
9
@yufu0413
yu fu
6 years
Great work done by @awj6674 Deep learning identifies prognostic factors auch as necrotic regions and infiltrating lymphocytes by learning from millions of H&E images tiles, using only patients’ overall survival.
@MoritzGerstung
Moritz Gerstung
6 years
Similarly the network finds prognostic associations in most cancer types that match and augment conventional grading and subtyping and points out prognostically relevant regions, such as necrosis, on each slide. 4/5
0
2
5
@MoritzGerstung
Moritz Gerstung
6 years
The network can predict a good range of genomic alterations, including whole genome duplications. From H&E-images alone. 2/5
1
15
37
@yufu0413
yu fu
6 years
Without spatial annotation of tumor lymphocyte, our method PC-CHiP (Pan-Cancer Computational HistoPathology) is able to automatically identify regions with lymphocytes for thousands of large H&E whole tissue slides. Great tool to assist pathologists!
@MoritzGerstung
Moritz Gerstung
6 years
It also finds a lot of associations in bulk transcriptome data, deconvolves the signal to find areas on each slide corresponding to molecular cell types such as tumour infiltrating lymphocytes. Entirely automated. 3/5
0
7
9
@MoritzGerstung
Moritz Gerstung
6 years
Hello world. Here’s something interesting: @yufu0413 from my lab trained a deep convolutional neural net in cancer histopathology *and* genomics using 14M images from 17k H&E slides across 28 cancer types. The outcome is stunning. 1/5 https://t.co/H3lGmoSKn8
15
325
678