Full papers
8-page papers (excluding references) following Springer LNCS format; supplementary materials (max 2 pages PDF) allowed for proofs or details; video or multimedia submissions permitted for relevant applications.
Machine Learning/CORE Unranked
UNSURE 2026
Uncertainty for Safe utilisation of Machine learning in Medical Imaging - Workshop at MICCAI 2026 - 8th edition
The UNSURE 2026 workshop, held as a satellite event of MICCAI 2026 in Strasbourg, focuses on uncertainty quantification and safety in medical imaging applications powered by machine learning. It aims to advance research on modeling uncertainty to ensure reliable clinical deployment, covering areas such as risk management, out-of-distribution detection, robustness to domain shifts, and validation of uncertainty estimates across medical image analysis and computer-aided intervention tasks.
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Research areas in scope
01
Uncertainty Quantification and Safety in Medical Imaging
Uncertainty quantification in any MIC or CAI applicationsRisk management of ML systems in clinical pipelinesOut-of-distribution and anomaly detectionDefending against hallucinations in enhancement tasks (e.g. super-resolution, reconstruction, modality translation)Robustness to domain shiftsMeasurement errorsModelling noise in data (e.g. labels, measurements, inter/intra-observer variability)Validation of uncertainty estimatesActive LearningConfidence boundsPosterior inference over point estimatesBayesian deep learningGraphical modelsGaussian processesCalibration of uncertainty measuresBayesian decision theory
Policies worth checking twice
- Submissions must be original and unpublished.
- Author names, affiliations, and identifying clues must be removed for blind review.
- The 8-page limit applies only to main content; references and acknowledgements may exceed this limit.
- Supplementary materials cannot include voiced-over slides.
- All accepted papers must submit a 6-minute presentation video.
- Dual submission to other conferences or journals is not permitted.
- Camera-ready submissions must be made by the deadline.
- Author lists cannot be changed after the submission deadline.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Official workshop siteProbabilistic Image Registration via Deep Multi-class Classification Characteriz
Last verified September 9, 2026