@cppape

Group leader @ Uni Goettingen applying Image Analysis and Deep Learning to large microscopy data in biology. Also @cppape.bsky.social

 Goettingen
user.informatik.uni-goettingen.de/~pape41/
Joined December 2015
  • Tweets1K
  • Following558
  • Followers1.3K
73Media
Constantin Pape@cppape
Jun 23
How can we use foundation models such as (micro)SAM to improve electron microscopy segmentation? Check out our preprint! We found substantial improvements for nucleus, mito, and neurite-segmentation using initialization and semi-supervised learning with foundation models.
1
9
47
3,544
Constantin Pape@cppape
Aug 12
AGI when ChatGPT can convert a PNG to SVG without creating ugly crop boundaries in each rasterized element.
141
Constantin Pape@cppape
Jul 16
Segmenting cells in microscopy is much easier these days thanks to foundation models. Can we use these models for other tasks, e.g. cell classification? See our latest work! We find big improvements for object and pixel classification compared to classical approaches.
valuation of different methods for pixel classification (top) and object classification (bottom) on the LIVECell dataset. Dark green bars show the F1-Score, which measures the classification / segmentation quality (higher is better), light green bars show the runtimes. Five different settings are compared for each task: Ilastik features + random forest (RF), microSAM embeddings + RF, SAM2 embeddings + RF, and uSAM, SAM2 + attentive probing (DeAP and ObAP). microSAM feature perform best for RF based methods, attentive probing outperforms RF based approaches, but at a much higher runtime.

ALT valuation of different methods for pixel classification (top) and object classification (bottom) on the LIVECell dataset. Dark green bars show the F1-Score, which measures the classification / segmentation quality (higher is better), light green bars show the runtimes. Five different settings are compared for each task: Ilastik features + random forest (RF), microSAM embeddings + RF, SAM2 embeddings + RF, and uSAM, SAM2 + attentive probing (DeAP and ObAP). microSAM feature perform best for RF based methods, attentive probing outperforms RF based approaches, but at a much higher runtime.

1
3
16
541
more replies
Constantin Pape@cppape
Jul 16
We compare foundation models (SAM and DINO variants) with classical features from ilastik and supervised learning. The FM features outperform classical approaches. Attentive probing is better than RF, but too slow for interactive use. See example results in the fig.
Results for object segmentation based on foundation model (FM) features with a random forest, compared with supervised learning baselines and classical features. Experiments over five different datasets covering cell classification in spatial proteomics (CRC, HBM), high-content macroscopic imaging for animal phenotyping (Planri), nucleus classification in histopathology (PanNuke), and cell line classification in phase-contrast microscopy (LIVECell). The extra plot repots runtimes for model training and inference.

ALT Results for object segmentation based on foundation model (FM) features with a random forest, compared with supervised learning baselines and classical features. Experiments over five different datasets covering cell classification in spatial proteomics (CRC, HBM), high-content macroscopic imaging for animal phenotyping (Planri), nucleus classification in histopathology (PanNuke), and cell line classification in phase-contrast microscopy (LIVECell). The extra plot repots runtimes for model training and inference.

1
2
116
Constantin Pape@cppape
Jul 16
This research will help us to build better tools for interactive microscopy analysis --- stay tuned for more! For more details: arxiv.org/abs/2603.19802 Thanks to Caro for driving this work and all other co-authors!

Evaluating Vision Foundation Models for Pixel and Object...

Deep learning underlies most modern approaches and tools in computer vision, including biomedical imaging. However, for interactive semantic segmentation (often called pixel classification in this...

arxiv.org
1
2
70
Constantin Pape@cppape
Jun 30
Want to learn about the latest trends and developments in bioimage analysis? Join us for the GLOBIAS workshop in Dortmund this November! I will be hosting a session on foundation models / AI. globias.org/activities/globi…

GloBIAS2026 Workshop Dortmund

GloBIAS 2026 Bioimage Analysts' Workshop Registration is open! Register with this link Registration is closed: We have filled all seats available. A waiting list has been created and can be accessed...

globias.org
2
8
435
Constantin Pape@cppape
Jun 23
How can we use foundation models such as (micro)SAM to improve electron microscopy segmentation? Check out our preprint! We found substantial improvements for nucleus, mito, and neurite-segmentation using initialization and semi-supervised learning with foundation models.
1
9
47
3,544
Constantin Pape@cppape
Jun 23
Learn more here: arxiv.org/abs/2606.21605 Thanks to Marei for leading this work!

$μ$Match: Foundation Models for Semi-supervised Learning and...

Vision foundation models have substantially advanced computer vision, enabling state-of-the-art performance in zero- and few-shot settings. They have been successfully applied to biomedical...

arxiv.org
2
178
Constantin Pape@cppape
Jun 15
We are so behind in Europe that our fanfic version of AI 2027 drops a year later and is only Europe 2031.
Tom Chivers@TomChivers
Jun 11
Here's a project I've been working on recently: a vision of what happens if Europe doesn't take AI seriously, inspired by AI 2027 europe2031.ai/

Europe 2031 — What getting AI wrong means for us

A five-year scenario about AI and Europe's impending slide into irrelevance, with a 2034 epilogue that describes how the collapse of the European model could have been prevented.

europe2031.ai
1
187
Constantin Pape@cppape
Jun 13
Did it actually answer someone what a mitochondrion is 😱😱😱
Anthropic
@AnthropicAI
Jun 13
The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees. The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance. Access to all other Claude models is not affected. We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible. Read our full statement: anthropic.com/news/fable-myt…

Statement on the directive to suspend Fable 5 access

The US government has issued an export control directive suspending all access to Fable 5 and Mythos 5 by foreign nationals inside or outside the US.

anthropic.com
1
181
Constantin Pape retweeted
Leo 🏴‍☠️
@leostera
May 22
i need someone at @OpenAI and @AnthropicAI to teach the models that while prototyping, backwards compatibility is just a bad idea
82
83
22
2,188
109,396
Constantin Pape@cppape
May 20
Sorry to break the news, but the job of twitter hypeman was automated a few years ago already.
Dr. Thomas Ichim
@exosome
May 20
Today we all lost our jobs..... Three Nature papers showing that scientists in the conventional sense are obsolete At least read the first one.... the AI replaced all things that the scientist does .... nature.com/articles/s41586-0…
5
729
🔬 Imaging Seminar on Vision Foundation Models for Microscopy with Prof. Constantin Pape (@cppape) from University of Göttingen 📅 Mar 30, 17:00 👉 Registration: lnkd.in/ej6iCuPA 💡 EPFL PhDs: Imaging Lunch w/ Prof. Pape 📅 Mar 31, 10:00 👉 lnkd.in/exemA5hW
4
10
544
Constantin Pape@cppape
Mar 5
Looking for a PhD position at the intersection of AI, imaging, and gene therapy? Apply for this position in my lab: tinyurl.com/2a2v6tvx Part of sfb1690.uni-goettingen.de/ that studies hearing, vision, and more. Plus, you can create pretty pictures like the one below :).
A cochlea imaged in light-sheet microscopy (right) with staining for spiral ganglion neurons (red) and inner hair cells (blue). You will develop AI-based methods to analyze these structures, for example via segmentation of the individual cells (right) that will support gene therapy development for hearing loss and a better overall understanding of the anatomy of hearing.

ALT A cochlea imaged in light-sheet microscopy (right) with staining for spiral ganglion neurons (red) and inner hair cells (blue). You will develop AI-based methods to analyze these structures, for example via segmentation of the individual cells (right) that will support gene therapy development for hearing loss and a better overall understanding of the anatomy of hearing.

2
8
2
40
2,610
Constantin Pape retweeted
Davis Blalock
@davisblalock
Mar 3
🚀 Today we’re releasing FlashOptim: better implementations of Adam, SGD, etc, that compute the same updates but save tons of memory. You can use it right now via `pip install flashoptim`. 🚀 arxiv.org/abs/2602.23349 A bunch of cool ideas make this possible: [1/n]
31
227
26
1,550
223,847
Constantin Pape retweeted
Timothy Fuqua@timothy_fuqua
Feb 10
I'm looking for a Swiss department to host me for an SNSF StG. I research how gene expression evolves and emerges by combining wet lab + computational work in a variety of model systems. More: timothyfuqua.com If your department might be a match, let’s chat!

Home | Timothy Fuqua

timothyfuqua.com
4
8
803
Constantin Pape@cppape
Jan 12
Retweeting as an AC: Please sumbit your CVPR reviews ;)
#CVPR2026@CVPR
Jan 12
Have you completed your reviews yet?
4
682
Constantin Pape retweeted
#CVPR2026@CVPR
Jan 11
The #CVPR2026 review deadline is fast approaching: January 12, 2026. If you face any difficulties, contact your Area Chair NOW! Approach your reviews the way you would expect your own to be handled, and consult the reviewer guidelines here: cvpr.thecvf.com/Conferences/…
1
9
2
34
31,276
Constantin Pape retweeted
Talia Ringer 🕊🪬@TaliaRinger
19 Nov 2025
OK can you guys please dedicate even half the energy you collectively use critiquing Empire of AI to also critique the massive work of harmful pseudoscience by Yud and Soares that was also a NYT Bestseller
15
16
3
155
18,666
Constantin Pape@cppape
18 Nov 2025
Preprint alert! CochleaNet, our framework for analyzing cochlear light-sheet data. It consists of three networks to segment spiral ganglion neurons, inner hair cells, and detect synapses. Rendering of a full cochlea below, find the preprint at doi.org/10.1101/2025.11.16.6…
1
4
386
more replies
Constantin Pape@cppape
18 Nov 2025
We comprehensively evaluate our method and use it to analyze intact mouse and gerbil cochleae, including SGN sub-types, and to validate (opto-)genetic therapies preclinically (see screenshot). CochleaNet is also applicable to lower-resolution data from commercial systems.
1
1
104
Constantin Pape@cppape
18 Nov 2025
This work was an excellent collaboration with the group of Tobias Moser, building on the cutting edge light-sheet microscopy of the Huisken lab. Special thanks to Lennart, Aleyna, Elisabeth, and Martin! If you want to use CochleaNet, check out github.com/computational-cel…

GitHub - computational-cell-analytics/cochlea-net: Deep learning-based image analysis for cochleae...

Deep learning-based image analysis for cochleae imaged in light-sheet microscopy - computational-cell-analytics/cochlea-net

github.com
2
194
Load more