CALM4GRAINS - Cross-domain Adaptation and Learning for Microscopy (2025-2028)

Ignacio Arganda-Carreras, Fadi Dornaika, Nagore Barrena, Urtzi Ayesta, Daniel Franco-Barranco, Lenka Backová, Aitor González-Marfil
Dec 1, 2025

The CALM4GRAINS subproject - Cross-domain Adaptation and Learning for Microscopy - is part of the coordinated GRAINS initiative (Green Responsible AI for Next-gen Sustainable Bio-Imaging). CALM4GRAINS focuses on improving model transferability across microscopy domains, laboratories, acquisition setups, and downstream tasks while reducing redundant computation.

Led at UPV/EHU by the Computer Vision and Pattern Discovery (CVPD) group, CALM4GRAINS contributes the adaptation and generalization layer of the coordinated project by leveraging reusable representations and foundation-model strategies tailored to biomedical microscopy.

🔬 Scientific Goals

  • Develop robust cross-domain adaptation and domain generalization methods for microscopy datasets with heterogeneous acquisition conditions.
  • Reuse representations across datasets and tasks to reduce retraining costs and improve computational and energy efficiency.
  • Adapt and benchmark foundation-model-based pipelines for microscopy segmentation, detection, and classification.
  • Improve reproducibility through shared protocols and interoperable software components compatible with open bioimage ecosystems.

🌐 Coordinated Project Context

CALM4GRAINS is Subproject 2 (SP2) within the coordinated project:

  • GRAINS: Green Responsible AI for Next-gen Sustainable Bio-Imaging
    • SP1: GAINS4GRAINS - Green Architectures for Imaging Networks and Sustainability
    • SP2: CALM4GRAINS - Cross-domain Adaptation and Learning for Microscopy
    • SP3: HEAT4GRAINS - Energy and hardware-efficient neural network training algorithms
    • SP4: REACT4GRAINS - Reconfigurable arithmetic and circuits for green AI networks

👥 Project Team

Principal Investigators

Research Team

Work Team / Collaborators

🕒 Timeline

  • Start date: 1 December 2025
  • End date: 30 November 2028
  • Duration: 36 months

💰 Funding

  • Call / Program: MICINN-IA (MICINN-IA25/03) - Research projects in the field of Artificial Intelligence (AIA 2025)
  • Funding body: State Research Agency (AEI), Spanish Ministry of Science, Innovation and Universities (MICIU)
  • External reference code: AIA2025-164165-C42
  • Official subproject title: Cross-domain Adaptation and Learning for Microscopy
  • Total funding granted: 328,500.00 EUR
Spanish State Agency for Research (AEI) Logo

📄 Project Documentation

  • Coordinated project: GRAINS (Green Responsible AI for Next-gen Sustainable Bio-Imaging)
  • Subproject reference: AIA2025-164165-C42

📬 Contacts

For project enquiries, please contact:

Biomedical Imaging Microscopy Domain Adaptation Domain Generalization Foundation Models Sustainable AI Green AI
Ignacio Arganda-Carreras
Ignacio Arganda-Carreras
Ikerbasque Research Associate Professor

My research interests include image processing, computer vision, and deep learning for biomedical applications.

Fadi Dornaika
Fadi Dornaika
Ikerbasque Research Professor

Ikerbasque Research Professor with expertise in computer vision, machine learning, and pattern recognition.

Nagore Barrena
Nagore Barrena
Assistant Professor

My research focuses on computer vision and deep learning, with applications to biomedical and natural images.

Daniel Franco-Barranco
Daniel Franco-Barranco
Postdoctoral Researcher

My primary focus is on the development of deep learning solutions for the segmentation of organelles in large-scale and multimodal electron microscopy images.

Lenka Backová
Lenka Backová
PhD Student

My research focuses on deep learning-based bioimage analysis and biophysical modeling of multicellular systems.

Aitor González-Marfil
Aitor González-Marfil
PhD Student

My research focuses on generalizable deep learning methods for computer vision.