Artificial intelligence (AI) has achieved or even exceeded human performance in many healthcare tasks, owing to the fast development of AI techniques and the growing scale of medical data. However, AI techniques are still far from being widely applied in healthcare practice. Real-world scenarios are far more complex, and AI is often faced with challenges in its credibility such as lack of explainability, generalization, fairness, privacy, etc. The development of trustworthy artificial intelligence for healthcare (TAI4H) is hence of great importance to enhance the trust and confidence of doctors and patients in using the related techniques. We aim to bring together researchers from interdisciplinary fields, including but not limited to machine learning, clinical research, and medical imaging, etc., to provide different perspectives on how to develop trustworthy AI algorithms to accelerate the landing of AI in healthcare.

Interested topics will include, but not be limited to:
Generalization to out-of-distribution samples.
Explainability of machine learning models in healthcare.
Reasoning, intervening, or causal inference.
Debiasing AI models from learning from shortcuts.
Fairness in medical imaging.
Uncertainty estimation of machine learning models and medical data.
Privacy-preserving AI for medical data.
Learning informative and discriminative features under weak annotations.
Human-machine cooperation (human-in-the-loop, active learning, etc.) in healthcare, such as medical image analysis.
Multi-modal fusion and learning, such as computed tomography (CT), magnetic resonance imaging (MRI), ultrasound, pathology, genetics, electronical healthcare records, etc.
Adversarial attack and defence in healthcare.
Benchmarks that quantify the trustworthiness of AI models in medical imaging tasks.
Foundation model pre-training and adaptation.

The goal of this TAI4H workshop is to bring together expertise from academia, clinic, and industry with an insightful vision of promoting trustworthy artificial intelligence for healthcare in terms of scalability, accountability, and explainability. The challenges to AI come from diverse perspectives in practice, and it is therefore of great importance to establish such an interdisciplinary platform to encourage sharing and discussion of ideas, implementation, data, labelling, benchmarks, experience, etc, and jointly advance the frontiers of trustworthy AI for healthcare.