AI4Life

AI4Life_Logo
Menu

Month: October 2025

Categories
News

In retrospect: AI4Life Challenges

In retrospect: AI4Life Challenges

AI4Life’s mission isn’t just to assist individual projects , it also seeks to benchmark and push the boundaries of AI methods for bioimage analysis. Over three years, the project has organised three public challenges, inviting the computational community to compete on core imaging tasks under shared datasets, metrics, and conditions.

Overview & evolution

First Challenge: unsupervised denoising

AI4Life launched the inaugural challenge focusing on unsupervised denoising of microscopy images. Rather than relying on paired “noisy/clean” image sets, participants were invited to apply algorithms that learn denoising from noisy data alone.

This setup reflects a practical constraint in microscopy: obtaining “clean” reference images is often difficult or impossible. The goal was to reduce noise while preserving delicate image features like edges, textures, and fine detail.

Four datasets were used, with two types of noise (structured and unstructured) represented. https://ai4life-mdc24.grand-challenge.org/ai4life-mdc24/

The 2024 AI4Life Denoising Challenge drew 104 registrants from 27 countries, resulting in 151 submissions across four leaderboards (each corresponding to a dataset/noise type).
Top results included combinations of algorithms like COSDD and N2V, which delivered strong performance across modules. 

Find out more: https://ai4life.eurobioimaging.eu/ai4life-denoising-challenge-2024-results/

Second & Third Challenges: transitioning to supervised denoising

Following the insights from 2024, the 2025 challenge shifts focus to supervised denoising, now combining noisy and clean images to train models. The change allows for more precise performance evaluation and potentially better denoising when ground truth is available. The 2025 edition moved to include supervised denoising (paired noisy/clean) and a specialised calcium-imaging track. 

The MDC25 and CIDC25 result pages (hosted on Grand-Challenge) hold the detailed leaderboards and entries; the CIDC25 platform remains accessible for benchmarking and late submissions:

These two challenges together drew 91 submissions from 13 countries across eight leaderboards, with more details at: 

https://ai4life.eurobioimaging.eu/ai4life-denoising-challenges-2025-results/

 

Why these challenges matter

  • Benchmarking: Standardised challenges let us compare methods fairly, across diverse datasets and noise types.
  • Broad community engagement: By opening up to anyone (not just project partners), AI4Life attracts fresh ideas and cross-pollination from adjacent fields.

From calls to conversations: reflections from our experts

Challenges are not just competitions; each involves substantial coordination, data curation, evaluation, and community outreach. To bring that human side forward, we recorded conversations with several experts who played key roles in designing, running, and evaluating these challenges.

Watch the full video and hear their stories: 

Categories
News

In retrospect: AI4Life Open Calls

In retrospect: AI4Life Open Calls

Over the past 3 years, AI4Life has launched a series of three annual Open Calls to support life scientists facing unmet image analysis needs. 

These calls invited scientists working with biological or microscopy images to propose analysis problems where deep learning makes a difference. Proposals were selected based not just on novelty, but on broader impact, reusability, and feasibility. 

What each Open Call offered
  • Technical support and collaboration: Project experts provided guidance on the development of analysis workflows, data improvement and deep learning solutions.
  • Consultation phase (from Open Call 2 onward): Before full project selection, top applicants engaged in a consultation phase to assess data readiness, existing methods’ suitability, and possible quick wins. In some cases, the consultation itself was enough to resolve issues.
  • Public dissemination: At the end of each project, the developed workflows, some subset of data, and trained models were made publicly available (e.g. via the BioImage Model Zoo or open archives).
  • Potential elevation into public challenges: Projects selected via the Open Calls addressing common needs could be turned into public challenges, inviting more involvement from the computational community to improve or benchmark solutions.
Evolution through three editions

First Open Call (Spring 2023) 

https://ai4life.eurobioimaging.eu/open-calls/first-open-call/ 

  • 72 proposals submitted from various life science domains, from cell and developmental biology to marine and plant research.
  • 67 aimed to improve existing workflows, while fewer proposed new AI-based methods.
  • The lack of annotated (“ground truth”) data was a recurrent bottleneck, revealing a widespread need for annotation support and AI-readiness training.

Second Open Call (Autumn 2023)

https://ai4life.eurobioimaging.eu/open-calls/second-open-call/ 

  • Introduced a two-phase process, beginning with a 1-hour consultation phase to assess project feasibility before full project support.
  • Project consultations increased interaction between applicants and experts, leading to better alignment of expectations and higher-quality collaborations.

Third Open Call (Spring 2024)

https://ai4life.eurobioimaging.eu/open-calls/third-open-call/

  • Received 28 applications across Europe, showing continuing demand despite a more targeted scope.
  • As in OC2, top applications underwent consultation before selected cases advanced to full technical support.

Across all calls, an international review committee has guided project selection by evaluating scientific impact, feasibility, reusability, and alignment with AI4Life’s capabilities.

Use Cases: real stories of impact

All projects selected through the Open Calls are showcased at the Use Cases page, illustrating how AI4Life support can turn complex bioimage tasks into reproducible, shared pipelines.

https://ai4life.eurobioimaging.eu/open-calls/use-cases 

These use cases show the diversity of problems addressed (2D/3D, multi-modal, multi-label) and how the solutions are made open for reuse.

 

From calls to conversations: reflections from our experts

Behind every open call was a dedicated group of experts guiding applicants, shaping proposals, and co-developing solutions. Watch the video below to hear their stories: