MICCAI 2026 Workshop

BRIDGE
Workshop

Bridging Regulatory Science and Medical Imaging Evaluation — from technical performance to real-world utility and clinical value.

DateSeptember 27 · Strasbourg, France

VenueMICCAI 2026

ProceedingsSpringer LNCS

From Performance to Patient Benefit

01

Evaluate It

Rigorous technical assessment of AI model performance

02

Measure Its Value

Does it add value, or genuinely improve outcomes and clinical decisions?

03

Deploy It

Safe, effective, and responsible clinical deployment

04

Monitor It

Post-market monitoring: drift detection, ongoing performance validation, and sustained safety and effectiveness

Regulatory-driven AI innovation. Every stage of AI development should align with safety standards, and every risk, every failure — must be measured, documented, and mitigated.

About the Workshop

Building Medical AI That Works — Safe, Effective, and Clinically Valuable

The field has focused heavily on developing state-of-the-art models, benchmarking on static datasets, and optimizing metrics that often do not reflect the clinical task. The result: AI models that perform well in research settings routinely fail to deliver meaningful benefit once deployed in clinical practice — and in some cases lead to serious consequences impacting patients and the healthcare system.

To build medical AI that works safely, effectively, and as intended — while adding real value to clinical practice — we must start with the end goal. What problem will this AI solve? What will it do for the patient? What value will it add to the clinician or health system? The end goal must be defined before the model is built.

We must also ensure that every stage of AI development has a mechanism to measure risk, that we choose the right metrics to evaluate those risks, and that every design and development decision aligns with safety and responsible AI principles.

"Design for translation. Evaluate what matters. Measure clinical value. Deploy responsibly. Monitor continuously."

What Makes BRIDGE Different

While many MICCAI workshops focus on technical innovation, state-of-the-art model development, and performance improvements on benchmark datasets, BRIDGE focuses on what comes next: how AI innovations should be evaluated, translated, and implemented so they can function safely and effectively in real world clinical settings. The workshop emphasizes the evaluation aspects of medical AI innovation, including whether models perform as intended, whether they are robust across populations and clinical environments, whether they provide evidence of clinical value, and whether they can be responsibly deployed and monitored after implementation.

BRIDGE brings together AI researchers, clinicians, regulatory scientists, health economists, industry leaders, and implementation experts to discuss how innovative medical imaging and medical AI technologies can move from research prototypes to clinically useful tools.

Explore

Get Involved

📄

Call for Papers

Submit original research, empirical studies, position papers, or negative results across all four pipeline stages.

View topics and dates

📅

Workshop Schedule

Program details coming soon. The workshop will bring together participants from academia, industry, and regulatory bodies.

See the program

👥

Organizers and Advisory Board

Meet the team of researchers, clinicians, and regulatory scientists behind BRIDGE.

Meet the team

Call for Papers

Submit Your Research

We invite original research papers, empirical studies, position papers, and negative results on topics related to the evaluation, clinical translation, regulation, and post-market monitoring of medical AI systems.

Submit on OpenReview

Topics of Interest

Research Topics

The BRIDGE track welcomes submissions on the evaluation and clinical translation of medical AI, including technical performance assessment, clinical utility, regulatory science, and post-deployment monitoring. Topics include but are not limited to:

Stage 01

AI Evaluation and Performance Assessment

Evaluation frameworks and methodologies for medical AI models in MIC and CAI
Performance metrics for diagnostic, prognostic, predictive, and decision-support AI applications, including metrics that better reflect downstream clinical actions and patient benefit
Evaluation of model robustness, generalizability, reliability, and failure modes across real-world operating conditions
Benchmark design, dataset curation, and limitations of static dataset evaluation for medical AI translation, including dataset representativeness, hidden biases, annotation quality
Evaluation of clinical value, including patient benefit, clinician decision support, workflow impact, care efficiency, resource utilization, etc.
Evaluation of large language models, multimodal foundation models, and generalist medical AI systems
Explainability and interpretability methods and their clinical relevance
Negative results and lessons learned from failed evaluations

Stage 02

Clinical Utility and Value Assessment

Methods for measuring clinical utility and decision impact of AI tools
Clinical validation study design for AI-based medical devices
Patient outcome metrics and how to link them to AI performance
Health economic evaluation and cost-effectiveness of AI systems
Studies reporting limited, negative, or unexpected clinical value
Identification of mismatches between technical metrics and clinical outcomes
Human factors and the effect of AI on clinician decision-making

Stage 03

Regulatory Science and Clinical Deployment

Regulatory frameworks for AI-based medical devices (FDA, EU MDR, and international)
Software as a Medical Device (SaMD) evaluation and approval pathways
Pre-market evaluation requirements and evidence standards for AI
Algorithmic bias, fairness, and equity in clinical AI deployment
Workflow integration and implementation challenges in clinical environments
Case studies of AI deployment: barriers, facilitators, and lessons learned
Ethics, transparency, and accountability in clinical AI systems

Stage 04

Post-Market Monitoring and Lifecycle Management

Post-market surveillance methods for deployed AI systems
Detection of performance degradation, data drift, and distribution shift
Frameworks for ongoing monitoring of safety and clinical effectiveness
Model updating, retraining, and version control under regulatory constraints
Continual learning systems and associated regulatory considerations
Incident reporting and corrective action processes for AI medical devices
Long-term real-world evidence generation for AI systems

Important Dates

Submission Deadlines

All deadlines are 23:59 Pacific Time.

1Jul 2026

Full Paper Deadline

23:59 Pacific Time

31Jul 2026

Notification of Acceptance

Authors notified by email

15Aug 2026

Camera-Ready Version

Final submission due

27Sep 2026

Workshop Day

MICCAI 2026 · Strasbourg, France

Format

Submission Format

All submissions must follow the official Springer LNCS format and be fully anonymized for double-blind peer review.

Official Springer LNCS format (Word and LaTeX templates available)
Up to 8 pages of main content
Up to 2 additional pages for references only
Double-blind peer review — fully anonymized submissions
Remove author names, affiliations, and self-identifying references
Submit via OpenReview

Accepted papers will be published in the MICCAI 2026 Workshops proceedings in Springer's Lecture Notes in Computer Science (LNCS) series.

Submit on OpenReview

Program

Workshop Schedule

Full program details are to be announced. The workshop will bring together participants from academia, industry, and regulatory bodies for a full day of presentations, discussions, and networking.

📅 September 27, 2026 · MICCAI 2026 · Strasbourg, France

📋

Program Coming Soon

The full workshop program is currently being finalized. The program will include keynote talks, oral and poster presentations, and a panel discussion featuring participants from academia, industry, and regulatory bodies.

🎤

Keynote Talks

Invited speakers from academia, industry, and regulatory agencies

📄

Paper Presentations

Oral and poster presentations of accepted research papers

💬

Panel Discussion

Cross-sector dialogue on regulatory-driven AI innovation

The Team

Organizers & Advisory Board

The BRIDGE workshop is organized by a team of researchers and scientists working at the intersection of medical AI, evaluation science, and regulatory research.

Organizers