AI4Life

Thank you to all our partners, contributors, and the global community who helped shape AI4Life!

All resources remain open and available for reuse, adaptation, and further development.

Use them. Cite them. Contribute back.

Now… What’s next?

ai4life.eurobioimaging.eu/whats-next-f...

A menagerie of deep-learning models | EMBL EMBL scientists collaborate to establish easy-to-use ‘Zoo’ of pre-trained AI models for microscopy image analysis.

@embl.org recently highlighted the BioImage Model Zoo in a Lab Matters feature on making deep-learning models more accessible to bioimaging researchers:
www.embl.org/news/lab-mat...

AI4Life also invested in collaboration beyond academia, working with industry and Life Science Research Infrastructures to align with real-world needs.

Read how AI4Life collaborated with industry:
ai4life.eurobioimaging.eu/in-retrospec...

Community co-creation was central to AI4Life.

Through Open Calls and Challenges, scientists and AI experts collaborated on real-world scientific problems.

Read about the Open Calls:
ai4life.eurobioimaging.eu/in-retrospec...
Read about the Challenges:
ai4life.eurobioimaging.eu/in-retrospec...

Open science was a core principle of AI4Life, not an afterthought.

The project ensured that tools, models, workflows, and knowledge remain open, reusable, and community-driven.

Read about Open Science in AI4Life:
ai4life.eurobioimaging.eu/in-retrospec...

What impact did AI4Life have?

From open AI models to sustainable services and infrastructure, the project laid foundations that will be reused well beyond its lifetime.

Read more about key results and impact:
ai4life.eurobioimaging.eu/in-retrospec...

As we wrap up the year, we’re reflecting on what AI4Life delivered and what continues beyond the project.
AI4Life concluded in August 2025, but its outcomes remain central to FAIR, accessible AI-powered bioimage analysis.
Here’s where to start:
www.eurobioimaging.eu/news/ai4life...

Imagine you’re a scientist studying how plants respond to environmental stress. You want to see how long-term drought conditions affect certain cell membrane structures, but doing it manually would take months of work, and you lack the skills to build a computer algorithm that could do it for you.

⏳ The countdown is on!
This is the final week to submit your entry to the AI4Life Challenges. Show what you can do and share your skills!
👇

⏳ Extra time unlocked!
The #AI4Life challenges deadline is now August 31 — more time to train, tweak, and triumph with your best AI models for bioimage analysis!

🚀 Take on the challenge:
1️⃣ Denoising: ai4life-mdc25.grand-challenge.org
2️⃣ Calcium Imaging: ai4life-cidc25.grand-challenge.org

#EUProject | #BioImaging & AI

➡️ BioImage Model Zoo by @ai4life.bsky.social is a platform for #AI models in biological imaging

➡️ To find out about the training courses and tools available to help you use the platform correctly for your area of study, visit: www.embrc.eu/project/ai4life/

🌊🧪🦑

In the meantime, we have more than 46 pretrained models from ⛴️ BiaPy available now in the 🦒 Bioimage Model Zoo 🫶 @ai4life.bsky.social

🔗 Have a look at: bioimage.io#/models

As usual, they are accesible from BiaPy's GUI 🖥️, Colab notebooks 📔 and our API 🤓

3rd challenge from AI4Life! We're inviting the community to design unsupervised models that preserve spatial & temporal patterns in neuronal data.

📅 Open till Aug 11, 2025
🚀 Join now: ai4life.eurobioimaging.eu/challenge-3/

- A postdoc to work on cancer extravasation (4-year project).
- A postdoc help in completing Filopodia-related lab projects (2-year).
- Multiple Marie Curie postdoc positions that will open in the fall.
- We will open a 5-year position to cover my teaching (70% research and 30% teaching).

🐾 The BioImage Model Zoo just got a major upgrade!

It has been fully redesigned, offering a new backend, model uploader, community curation tools, and integrated browser-based test runs powered by BioEngine. 🧪

Explore now 👉 bioimage.io

AI4Life Microscopy Denoising Challenge - Grand Challenge Wellcome to AI4Life-MDC24! In this challenge, we want to focus on an unsupervised denoising of microscopy images. By participating, researchers can contribute to a critical area of scientific research...

If unsupervised approaches are your thing, also check out the unsupervised denoising challenge and submit your solutions (the challenge is from last year, but we are still taking entries from everyone who is interested).

👉 ai4life-mdc24.grand-challenge.org

The 2nd AI4Life Challenge is live!
Calling the AI & bioimaging community to tackle a key microscopy challenge: removing noise while preserving detail.

📦 Paired noisy/clean datasets
📈 Ground-truth evaluation
🧠 DL focus

Build, test, compete 👉 ai4life.eurobioimaging.eu/challenge-2/

Ever wished your AI model could test itself before you download it? Now it can.

AI4Life has a new model evaluation platform that connects models from the BioImage Model Zoo to real data in the @bioimagearchive.bsky.social.

Details here 👉 ai4life.eurobioimaging.eu/the-ai4life-...

@ai4life.bsky.social & @eu-openscreen.bsky.social link highlighted in the isciii eu infrastructures for health event

The AI4Life Community Event comes to a close. We wrapped up with a big-picture discussion around the sustainability of services and the role that AI4Life plays in the EOSC.
Thank you to everyone who joined us in Helsinki and online! 🦒
Stay connected 💚

📘 Where science meets policy: Stefan Liebler (European Commission 🇪🇺) shares insights on Open Science, AI, EOSC, and how projects like AI4Life support the European research strategy. But also, he showcased notebooks running on the EOSC EU node! 🤖

☁️ Interoperability in action: @ai4eosc.eu’s AI4Life cloud marketplace lets researchers run Bioimage.IO models. A true success story of aligning European infrastructure for science. Thanks to our friends at @ai4eosc.eu for making this possible!

🌐 Shoutout to @globias.bsky.social, one of our global partners in promoting community-driven AI presented by @mycroscopy.bsky.social

🎥 Did you catch our joint seminar series? Don't forget to register for the upcoming ones!
bit.ly/registration...

Partner Research infrastructures in AI4Life have been instrumental in the reach of use cases gathered at the open calls. We heard today about some successful projects!

@embrc-eu.bsky.social, EU-OPENSCREEN, EMPHASIS and @instruct-eric.bsky.social

Industry collaboration is key to real-world impact. @zeiss-microscopy.bsky.social is a contributor to CAREamics, one of the BioImage.IO community partners. Also, Bioimage.IO models can be applied right after image acquisition with ZEN API. We need more of that!

💬 After lunch, we’re back with the AI4Life Community: tool builders, research infrastructures, and industry voices. Community tools like @biapyx.bsky.social and SpotMAX show how AI4Life partners are building bridges between developers and end users.

Yay! Well done @danifranco93.bsky.social 🤗 and long life to @ai4life.bsky.social 🤩

💡Success stories from AI4Life: From the BioImage.IO models to @bioimagearchive.bsky.social’s AI-ready datasets, and community resources like the Helpdesk, Open Calls, and Challenges — see how it all connects in one ecosystem!

What an amazing two days it’s been! Follow @ai4life.bsky.social to learn more about the amazing talks on AI in life science and how it’s shaping the future of life science research.