Merlin (Stanford abdominal CT vision-language model)
Merlin is a model trained on 15,000 CT scans with their reports that can find and describe hundreds of findings.
Merlin (2024) is a 3D vision-language model trained on 6M images and 6M EHR codes from abdominal CTs; supports zero-shot finding classification and report generation.
How it works
3D image encoder aligned with report text and structured codes.
- Report-supervised 3D learning
- Single institution
- Abdomen only
Latest papers
topQuery for this technology: (TITLE:"Merlin" OR ABSTRACT:"Merlin" OR TITLE:"Stanford abdominal CT vision-language model" OR ABSTRACT:"Stanford abdominal CT vision-language model") AND (cancer OR tumor OR tumour OR oncology OR carcinoma OR lymphoma OR leukemia OR leukaemia OR myeloma OR sarcoma OR melanoma OR glioma). Results are unfiltered search hits about Merlin (Stanford abdominal CT vision-language model), not a curated reading list.
Pages like this
not linked directly; found by shared links- TechnologyCT-FM (whole-body CT foundation model)
Shares CT (computed tomography), AI in radiology and the tags foundation-model, radiology.
- TechnologyAidoc CARE (clinical radiology foundation model)
Shares AI in radiology and the tags foundation-model, radiology.
- TechnologyRadFM (generalist radiology foundation model)
Shares AI in radiology and the tags foundation-model, radiology.
- TechnologyMedSAM / SAM-Med3D (segment anything for medicine)
Shares AI in radiology and the tags foundation-model, radiology.
- TechnologyGenePT
Shares Stanford Health Care / Stanford Cancer Institute and the tag foundation-model.
- TechnologySybil (MIT/MGH lung cancer risk from CT)
Shares CT (computed tomography), AI in radiology and the tag radiology.
- IdeaAI malignancy scores to end repeat scans and biopsies for benign lung nodules
Shares CT (computed tomography), AI in radiology.
- TechnologyMUSK (Stanford, vision-language pathology)
Shares Stanford Health Care / Stanford Cancer Institute and the tag foundation-model.