from __future__ import annotations
__all__ = ["UMLSNormAttribute"]
import dataclasses
from typing import Any, Dict, List, Optional
from typing_extensions import Self
from medkit.core import dict_conv
from medkit.core.text import EntityNormAttribute
[docs]@dataclasses.dataclass(init=False)
class UMLSNormAttribute(EntityNormAttribute):
"""
Normalization attribute linking an entity to a CUI in the UMLS knowledge base
Attributes
----------
uid:
Identifier of the attribute
label:
The attribute label, always set to :attr:`EntityNormAttribute.LABEL
<.core.text.EntityNormAttribute.LABEL>`
value:
CUI prefixed with "umls:" (ex: "umls:C0011849")
kb_name:
Name of the knowledge base. Always "umls"
kb_id:
CUI (Concept Unique Identifier) to which the annotation should be linked
cui:
Convenience alias of `kb_id`
kb_version:
Version of the UMLS database (ex: "202AB")
umls_version:
Convenience alias of `kb_version`
term:
Optional normalized version of the entity text
score:
Optional score reflecting confidence of this link
sem_types:
Optional IDs of semantic types of the CUI (ex: ["T047"])
metadata:
Metadata of the attribute
"""
sem_types: Optional[List[str]] = None
def __init__(
self,
cui: str,
umls_version: str,
term: Optional[str] = None,
score: Optional[float] = None,
sem_types: Optional[List[str]] = None,
metadata: Optional[Dict[str, Any]] = None,
uid: Optional[str] = None,
):
super().__init__(
kb_name="umls",
kb_id=cui,
kb_version=umls_version,
term=term,
score=score,
metadata=metadata,
uid=uid,
)
self.sem_types = sem_types
@property
def cui(self):
return self.kb_id
@property
def umls_version(self):
return self.kb_version
def to_dict(self) -> Dict[str, Any]:
norm_dict = dict(
uid=self.uid,
cui=self.cui,
umls_version=self.umls_version,
term=self.term,
score=self.score,
sem_types=self.sem_types,
metadata=self.metadata,
)
dict_conv.add_class_name_to_data_dict(self, norm_dict)
return norm_dict
[docs] @classmethod
def from_dict(cls, data: Dict[str, Any]) -> Self:
return cls(
uid=data["uid"],
cui=data["cui"],
umls_version=data["umls_version"],
term=data["term"],
score=data["score"],
sem_types=data["sem_types"],
metadata=data["metadata"],
)
