BoneCoT: multicentre validation of a whole-body skeleton foundation model for bone metastases guided by clinician-derived chain of thought.
AI model detects bone cancer spread on imaging with accuracy matching expert radiologists, potentially streamlining how oncologists stage patients.
BoneCoT is a whole-body skeleton foundation model that uses clinician-derived chain-of-thought reasoning to automatically detect and characterise bone metastases on imaging, validated across multiple centres. The model achieved performance aligned with expert radiological interpretation, demonstrating the potential of AI foundation models for automated oncologic bone staging in clinical workflows.
What the study was
- Study design
- Multicentre AI foundation model validation study
- Population
- Cancer patients with bone metastases on whole-body imaging
- Category
- Diagnostics
- Maturity
- Exploratory
- Journal
- Nat Biomed Eng
Why it surfaced
Top-tier journal (Nature Biomedical Engineering), multicentre-validated AI foundation model for bone metastasis staging; bridges AI/ML diagnostics and precision oncology.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.