[ISBI 2026 CFP] Special Session on Privacy-Aware, Data-Efficient AI via Personalized Incremental and Federated Learning in Healthcare

[Raffaele Mineo]

Dear colleagues,

we invite you to submit your work to the ISBI 2026 Special Session “Privacy-Aware, Data-Efficient AI via Personalized Incremental and Federated Learning in Healthcare”.

This session is part of the 2026 IEEE International Symposium on Biomedical Imaging (ISBI 2026), to be held in London, UK, on April 8-11, 2026.

The session tackles a very concrete problem in medical imaging: data often cannot be centralized because of privacy, governance and regulation, and even large hospitals rarely reach the scale needed for foundation-style training. Combining personalized incremental/continual learning with federated learning is a realistic path to make AI adaptive, distributed and clinically deployable across hospitals.

• Submission deadline: 14 November 2025
• Submission site: https://edas.info/N34160
• Session info: https://biomedicalimaging.org/2026/special-sessions/
• Conference: IEEE ISBI 2026, April 8–11, 2026, London, UK

 

Scope and Topics

We welcome methodological, applied and position/review papers. Submissions are encouraged (but not limited) to:

• Federated learning (FL) in medical imaging: cross-silo FL for hospitals, non-IID/heterogeneous sites and scanners, communication-efficient FL, secure aggregation, governance, evidence from real hospital federations
• Continual / incremental / lifelong learning (CL/IL) for clinical data: rehearsal-free and privacy-aware updates, stability–plasticity, handling protocol/scanner drift, practical medical CL evaluation
• Personalized models over time: parameter-efficient, streaming or on-device adaptation without centralizing data
• Privacy, safety and compliance: differential privacy, confidential computing, de-identification, alignment with EU/US regulatory expectations
• Evaluation and benchmarking for FL + CL: realistic longitudinal and multi-center protocols, metrics beyond accuracy (calibration, robustness, fairness, uncertainty), lessons from decentralized challenges
• Foundation-model pragmatics with limited data: federated adaptation, multi-center incremental specialization
• Clinical use-cases: radiography/CT/MRI, interventional imaging (e.g. angiography), oncology and radiotherapy, rare-disease/low-prevalence tasks
• Explicitly welcome:
• papers on medical continual/incremental learning only
• papers on medical federated learning only
• survey, review or discussion papers on these themes and their clinical translation

 

Submission Rules (ISBI 2026)

ISBI 2026 uses a single-blind reviewing process for full-length papers. Main rules (see full details at https://biomedicalimaging.org/2026/initial-author-instructions/ ):

• Full-length papers are limited to 4 pages (IEEE format).
• A 5th page is allowed ONLY for non-technical content:
1. Ethical compliance
2. Acknowledgments and conflict of interest
3. References
• All technical content (text, figures, tables) must be within the first 4 pages. If technical content appears on page 5, the paper will be rejected.
• Maximum length is 5 pages. Papers longer than 5 pages will be rejected.
• A $200 fee applies if the 5th page is used. This is paid at registration.
• Manuscript templates are available on the ISBI 2026 website.
• Accepted papers will appear in the ISBI 2026 proceedings and follow the same scientific standards as regular papers.

Please submit through EDAS ( https://edas.info/N34160 ) and select the special session “Privacy-Aware, Data-Efficient AI via Personalized Incremental and Federated Learning in Healthcare”.

 

We would appreciate it if you could forward this CFP to colleagues working on medical AI, distributed learning, continual learning, privacy-preserving methods, and clinical imaging.

Best Regards
On behalf of the organizers,

Raffaele Mineo (University Campus Bio-Medico of Rome)

Simone Palazzo (University of Catania)

Giovanni Bellitto (University of Catania)

Amelia Sorrenti (University Campus Bio-Medico of Rome)

Federica Proietto Salanitri (University of Catania)

Concetto Spampinato (University of Catania)