Biomedical NER Families¶
OpenMed's manifest-backed model registry groups trained named-entity recognition checkpoints into the families below. Entity types, available size categories, and confidence defaults reflect the current registry; the documentation coherence test prevents this catalog from drifting as the manifest changes.
The linked zero-shot domains come from available_domains() and open the generated Clinical Domain Label Catalog. See the Zero-shot NER Toolkit for label-resolution behavior and custom domain maps.
Recommended confidence values are starting thresholds, not universal operating points. Validate them on representative local data before production use. This catalog describes extraction coverage only and is not clinical guidance.
Medical¶
- Entity types:
CANCER,CELL,CHEM,CONDITION,DISEASE,DNA,DRUG,GENE,GENE_OR_GENE_PRODUCT,MEDICATION,ORGANISM,PATHOLOGY,PROTEIN,RNA,SIMPLE_CHEMICAL,SPECIES - Available size categories:
Tiny,Small,Medium,Large,XLarge,Unknown - Recommended confidence:
0.60 - Zero-shot domain:
biomedical
Anatomy¶
- Entity types:
ANATOMY,ORGAN,TISSUE - Available size categories:
Tiny,Small,Medium,Large,XLarge - Recommended confidence:
0.60 - Zero-shot domain:
anatomy
Hematology¶
- Entity types:
CANCER,CELL,CL,DISEASE - Available size categories:
Tiny,Medium,Large,XLarge - Recommended confidence:
0.65 - Zero-shot domain:
hematology
Chemical¶
- Entity types:
CHEM,CHEMICAL,DRUG,MEDICATION,SIMPLE_CHEMICAL - Available size categories:
Tiny,Small,Medium,Large,XLarge - Recommended confidence:
0.60 - Zero-shot domain:
chemical
Disease¶
- Entity types:
CONDITION,DISEASE,PATHOLOGY - Available size categories:
Tiny,Medium,Large,XLarge - Recommended confidence:
0.60 - Zero-shot domain:
disease
Genomics¶
- Entity types:
CELL,CELL_LINE,CELL_TYPE,DNA,GENE,GENE_OR_GENE_PRODUCT,PROTEIN,RNA - Available size categories:
Tiny,Medium,Large,XLarge - Recommended confidence:
0.65 - Zero-shot domain:
genomics
Oncology¶
- Entity types:
AMINO_ACID,ANATOMICAL_SYSTEM,ANATOMY,CANCER,CELL,CELLULAR_COMPONENT,CHEM,CHEMICAL,CONDITION,DEVELOPING_ANATOMICAL_STRUCTURE,DISEASE,GENE_OR_GENE_PRODUCT,IMMATERIAL_ANATOMICAL_ENTITY,MULTI_TISSUE_STRUCTURE,ORGAN,ORGANISM,ORGANISM_SUBDIVISION,ORGANISM_SUBSTANCE,PATHOLOGICAL_FORMATION,PATHOLOGY,PROTEIN,SIMPLE_CHEMICAL,SPECIES,TISSUE - Available size categories:
Tiny,Small,Medium,Large,XLarge - Recommended confidence:
0.65 - Zero-shot domain:
oncology
Species¶
- Entity types:
ORGANISM,SPECIES - Available size categories:
Tiny,Medium,Large,XLarge - Recommended confidence:
0.60 - Zero-shot domain:
species
Pathology¶
- Entity types:
CONDITION,DISEASE,PATHOLOGY - Available size categories:
Tiny,Medium,Large,XLarge - Recommended confidence:
0.60 - Zero-shot domain:
pathology
Pharmaceutical¶
- Entity types:
CHEM,CHEMICAL,DRUG,MEDICATION - Available size categories:
Tiny,Medium,Large,XLarge - Recommended confidence:
0.65 - Zero-shot domain:
pharmaceutical
Protein¶
- Entity types:
GENE_OR_GENE_PRODUCT,PROTEIN,PROTEIN_COMPLEX,PROTEIN_ENUM,PROTEIN_FAMILIY_OR_GROUP,PROTEIN_VARIANT - Available size categories:
Tiny,Medium,Large,XLarge - Recommended confidence:
0.60 - Zero-shot domain:
protein