Workshops

Thursday | April 9, 2026 | 8:00 – 17:30

Artificial Intelligence and Machine Learning (AI/ML)-enabled medical devices are advancing rapidly to address the evolving needs of patients, clinicians, and manufacturers in the MedTech industry. However, the pace of technological innovation has outstripped the development of evaluation methods in some instances, creating uncertainty for the evaluation of innovative technology.

The workshop aims to introduce regulatory science on AI-enabled devices to the ISBI community, with a particular focus on the assessment of medical video AI and uncertainty quantification. These areas present unique challenges, including frame-to-frame variability, motion artifacts, and temporal consistency. We aim to foster dialogue among researchers, clinicians, and regulators to discuss technical and regulatory challenges and help to reduce the gap between development of novel AI technologies and their evaluation and adoption. We will discuss the development of regulatory science tools, testing methods, and metrics for assessing AI-assisted devices.
While this workshop will focus on medical video AI systems, assessment many of the concepts generalize to other device types.

Thursday | April 9, 2026 | 8:00 – 17:30

Understanding the developing pediatric brain is critical for advancing pediatric healthcare and neuroscience. Early life is a period of rapid brain development and vulnerability, during which structural and functional brain alterations are closely linked to long-term cognitive and motor outcomes. Yet, studying the pediatric brain remains highly challenging due to dynamic morphology, limited data availability, and the lack of specialized analysis tools.

Despite progress in neuroimaging, many methods are optimized for adult brains and do not generalize well to pediatric populations. Furthermore, integration of multimodal data—including imaging, electronic health records, radiology reports, genetics, and social determinants—has been limited. This workshop will provide a dedicated forum for advancing pediatric brain data analysis.

Over the past three years, we have organized the international BONBID-HIE challenges on International Conference on Medical Image Computing and computer Assisted Intervention (MICCAI) 2023 and 2024, which focused on neonatal brain injury hypoxic-ischemic encephalopathy (HIE, affecting 1–5 per 1,000 newborns) and attracted more than 3,200 dataset downloads and over 300 participants from over 20 countries worldwide. Building on this foundation, we hope this workshop will broaden the scope to encompass infant and ediatric brain imaging more generally, addressing diverse conditions, multimodal data integration, and clinically relevant algorithms. In doing so, we aim to unite clinicians, engineers, and scientists to develop machine learning models with strong clinical relevance, promoting reproducibility, explainability, and clinical adoption.

This workshop aligns closely with ISBI’s emphasis on biomedical imaging and signal processing by fostering innovation in imaging methods, quantitative biomarkers, and multimodal data fusion for pediatric brain health. By bridging imaging physics, image analysis, and clinical interpretation, the workshop will contribute directly to ISBI’s goal of advancing translational research in biomedical imaging.

Friday | April 10, 2026 | 8:00 – 11:30

Large models have revolutionized computer vision and natural language processing, yet their application to surgery and interventional science is still in its infancy. With the unique potential to enhance surgical precision, efficiency, and patient safety, this emerging field offers exciting opportunities for innovation. Our workshop, Large Models Meet Surgical Data Science, will explore how large models can advance the analysis of surgical and interventional data, encompassing pre-operative imaging, intraoperative signals, robotic kinematics, tool presence cues, and electronic health records.

Despite this promise, major challenges remain. Surgical data is heterogeneous and often scarce, with limited annotated datasets to support model training. Current large model architectures are rarely tailored to the complexities of surgical workflows, and stringent requirements for privacy, security, and seamless clinical integration create further barriers.
This workshop will be the first dedicated forum to address these challenges at scale. It will feature keynote talks, technical presentations, and discussions that focus on building large-scale surgical datasets, designing robust and domain-specific models, and examining the ethical and regulatory implications of deploying AI in the operating room. By connecting experts across AI, medicine, and industry, the event aims to define a new standard for surgical data science and foster an inclusive, collaborative community.

Thursday | April 9, 2026 | 15:00 – 17:30

The rapid emergence of foundation AI models, large-scale pre-trained architectures such as vision transformers, diffusion models, and multimodal encoders, has ushered in a transformative era in medical image analysis. Leveraging massive natural and/or medical datasets, these models exhibit strong zero-shot and few-shot learning capabilities, cross-modality generalization, and the potential to unify diverse imaging tasks under a common representation framework. This session will critically examine the evolving role of foundation models in medical imaging, with applications spanning classification, segmentation, registration, and automated report generation.

Invited talks and expert panels will highlight both groundbreaking applications and open challenges. Key discussion topics include: (1) architectural advances and pretraining paradigms (e.g., self-supervised, contrastive, or generative learning); (2) domain adaptation across imaging modalities such as CT, MRI, ultrasound, and hybrid systems like PET/CT or PET/MRI; (3) interpretability, uncertainty quantification, and failure modes in safety-critical settings; and (4) issues of data curation, privacy, and regulatory compliance when deploying foundation models in clinical environments. The session will also emphasize the importance of interdisciplinary collaboration, bridging computer vision, medical imaging, medical physics, and clinical practice, and explore how foundation models may serve as the backbone for digital twin frameworks, image-guided interventions, or real-time adaptive therapies. Special focus will be given to risks including model over-reliance, hallucinations, and domain shifts, with calls for robust validation, benchmarking, and ethical deployment.

Saturday | April 11, 2026 | 8:00 – 11:30

The POCUS–AI Workshop highlights recent advances in artificial intelligence (AI) for point-of-care ultrasound (POCUS). Pocket-sized POCUS devices are increasingly used across diverse clinical settings due to their portability, safety, and ability to provide real-time imaging. However, widespread adoption remains limited by operator dependence and variability in interpretation. AI offers promising solutions by enabling more reliable, automated analysis through techniques such as segmentation, classification, and video summarization, as well as emerging clinical applications. This workshop will showcase cutting-edge research, foster cross-disciplinary collaboration, and bring together experts from academia, industry, and clinical practice to advance the future of POCUS–AI.