Bridging Computer Vision State-of-the-Art with Clinical Impact
See you in Arizona!
Pixels to Patients (P2P-CV) is a full-day workshop dedicated to bridging the gap between computer vision research and its safe, effective deployment in real-world clinical practice.
Medical imaging is among the most socially impactful domains of computer vision, yet a persistent gap remains between research breakthroughs and their safe, effective use in real-world clinics. WACV, with its strong tradition in applied computer vision, is the ideal venue to address this translation gap.
This workshop positions healthcare as a case study for broader CV challenges that arise whenever algorithms move from the lab to deployment. Topics such as domain generalization, fairness and bias mitigation, foundation models, trustworthy and explainable AI, and human–AI collaboration are central not only to medical imaging but also to autonomous driving, robotics, and other safety-critical applications.
Cutting-edge computer vision methods for medical imaging
Real-world deployment and patient care outcomes
Bridging researchers, clinicians, and industry
Fairness, explainability, and safety in healthcare
We invite original research contributions that not only present technical innovations but also critically engage with the challenges of translating computer vision from research into clinical practice.
Pre-training, fine-tuning, and adaptation for data-scarce medical imaging environments
Multimodal learning for clinical decision support and automated reporting
Improving robustness across scanners, sites, and patient populations
Validation, monitoring, and continual learning in practice
Ensuring safety, transparency, and interpretability
Auditing and mitigating bias for diverse populations
Active, weak, and self-supervised learning approaches
Using vision models for novel biomarker discovery
Collaborative interfaces and human-in-the-loop learning
Length: Up to 8 pages (excluding references)
Type: Archival - Published in WACV 2026 Workshop Proceedings
Purpose: Complete, original research with clinical relevance discussions
Learn More →Length: 2-4 pages (excluding references)
Type: Non-archival
Purpose: Preliminary work, ongoing research, or negative results
Learn More →Length: 2 slides maximum
Type: Non-archival
Purpose: Clinicians/industry present clinical challenges seeking technical solutions
Learn More →All deadlines are 23:59 Anywhere on Earth (AoE)
October 2025
December 8, 2025
January 12, 2026
January 30, 2026
March 6, 2026
Full-day schedule — March 6, 2026
Pixels to Patients: Dissolving the Immiscible
Mayo ClinicAccelerating ML Development For Healthcare through Open Weight Foundation Models
Google Research (Health AI)Intelligent Imaging Pipelines for Cardiology: From Algorithms to Clinical Insights
Mayo ClinicPoster elevator pitches on WS stage
AI in Digital Pathology
Harvard Medical SchoolDeployment in the Real World
Mayo Clinic + Canon MedicalM.D., Ph.D.
Mayo Clinic
Professor of Radiology and Director of the Mayo Clinic AI Lab. His work focuses on quantitative imaging and computer-aided diagnosis, developing AI algorithms for disease detection, prognosis, and prediction of molecular markers.
Ph.D.
Google Research (Health AI)
Research lead for foundation models and multimodal learning in healthcare. His work includes the development of open weight foundation models and advancing the application of large-scale models in medicine.
M.D.
Mayo Clinic — Department of Cardiology
Develops intelligent imaging pipelines linking algorithmic insights to real-world outcomes in cardiology, bridging the gap between advanced CV techniques and clinical practice.
Ph.D.
Harvard Medical School & Brigham and Women's Hospital
Associate Professor of Pathology and computational pathologist pioneering AI methods for digital pathology and precision medicine. Leads research on foundation models for histopathology, multimodal learning, and weakly-supervised learning for cancer diagnosis and prognosis.
M.D.
Mayo Clinic AI Lab
Postdoctoral Research Fellow specializing in medical computer vision, segmentation, radiology vision-language models, and active learning. Previously Senior AI/ML Engineer at Turkcell.
[email protected]M.D., Ph.D.
Mayo Clinic AI Lab
Professor of Radiology and Director of the Mayo Clinic AI Lab. Pioneer in quantitative imaging and computer-aided diagnosis.
[email protected]M.D., M.P.H., M.H.P.E.
Yale University
Radiology resident with extensive experience in full-stack development and machine learning, focusing on the intersection of radiology, AI, and medical education.
[email protected]M.D., M.P.H., M.H.P.E.
Yale University
Diagnostic Radiology resident applying machine learning to radiology imaging and clinical workflow optimization, exploring generative AI in education.
[email protected]M.D., M.P.H.
Mayo Clinic AI Lab
Postdoctoral Research Fellow focusing on deep learning analysis of cardiac imaging and textual medical documents.
[email protected]Direct dialogue between CV researchers, clinicians, and industry leaders on practical deployment challenges
Clear understanding of methodological, infrastructural, and regulatory obstacles to safe AI deployment
Problem-Pitch sessions catalyzing partnerships between those with clinical problems and technical solutions
Accepted papers published in official WACV 2026 workshop proceedings
Focus on fairness, diversity, and accessibility across all deployment contexts
Insights applicable to autonomous driving, robotics, and other safety-critical AI applications
Format: Full-Day Workshop
Expected Audience: 50-90 participants
Publication: Official WACV 2026 Workshop Proceedings
Review: Double-blind peer review