status mapping as defaultdict in disorders

Author: Thomzoy

edsnlp/pipes/ner/behaviors/alcohol/alcohol.py

@@ -1,5 +1,6 @@
 """`eds.alcohol` pipeline"""
 
+from collections import defaultdict
 from typing import Any, Dict, List, Optional, Union
 
 from spacy.tokens import Doc, Span
@@ -99,10 +100,10 @@ def __init__(
             nlp=nlp,
             name=name,
             patterns=patterns,
-            detailed_status_mapping={
-                1: None,
-                2: "ABSTINENCE",
-            },
+            detailed_status_mapping=defaultdict(
+                lambda: None,
+                {2: "ABSTINENCE"},
+            ),
             label=label,
             span_setter=span_setter,
             include_assigned=True,
@@ -118,8 +119,8 @@ def process(self, doc: Doc) -> List[Span]:
                 stopped = self.negation.process(span)
                 if not any(stopped_token.negation for stopped_token in stopped.tokens):
                     span._.status = 2
-
-            if "zero_after" in span._.assigned.keys():
-                span._.negation = True
+            else:
+                if "zero_after" in span._.assigned.keys():
+                    span._.negation = True
 
             yield span

edsnlp/pipes/ner/behaviors/alcohol/patterns.py

@@ -3,7 +3,7 @@
     regex=[
         r"\balco[ol]",
         r"\bethyl",
-        r"(?<!(25.?)|(sevrage)).?\boh\b",
+        r"(?<!(25.{0,10}))\boh\b",
         r"exogenose",
         r"delirium.tremens",
     ],

edsnlp/pipes/ner/disorders/base.py

@@ -1,4 +1,5 @@
 import re
+from collections import defaultdict
 from typing import Any, Dict, List, Union
 
 from spacy.tokens import Doc, Span
@@ -48,9 +49,9 @@ def __init__(
         include_assigned: bool = True,
         ignore_excluded: bool = True,
         ignore_space_tokens: bool = True,
-        detailed_status_mapping: Dict[int, Union[str, None]] = {
-            1: None,
-        },
+        detailed_status_mapping: Dict[Union[int, None], Union[str, None]] = defaultdict(
+            lambda: None
+        ),
         alignment_mode: str = "expand",
         regex_flags: Union[re.RegexFlag, int] = re.S,
         span_setter: SpanSetterArg,

edsnlp/pipes/ner/disorders/diabetes/diabetes.py

@@ -1,5 +1,6 @@
 """`eds.diabetes` pipeline"""
 
+from collections import defaultdict
 from typing import Any, Dict, List, Optional, Union
 
 from spacy.tokens import Doc, Span
@@ -102,10 +103,10 @@ def __init__(
             nlp=nlp,
             name=name,
             patterns=patterns,
-            detailed_status_mapping={
-                1: "WITHOUT_COMPLICATION",
-                2: "WITH_COMPLICATION",
-            },
+            detailed_status_mapping=defaultdict(
+                lambda: "WITHOUT_COMPLICATION",
+                {2: "WITH_COMPLICATION"},
+            ),
             label=label,
             span_setter=span_setter,
         )

edsnlp/pipes/ner/disorders/liver_disease/liver_disease.py

@@ -1,5 +1,6 @@
 """`eds.liver_disease` pipeline"""
 
+from collections import defaultdict
 from typing import Any, Dict, List, Optional, Union
 
 from spacy.tokens import Doc, Span
@@ -94,10 +95,10 @@ def __init__(
             nlp=nlp,
             name=name,
             patterns=patterns,
-            detailed_status_mapping={
-                1: "MILD",
-                2: "MODERATE_TO_SEVERE",
-            },
+            detailed_status_mapping=defaultdict(
+                lambda: "MILD",
+                {2: "MODERATE_TO_SEVERE"},
+            ),
             label=label,
             span_setter=span_setter,
         )

edsnlp/pipes/ner/disorders/solid_tumor/solid_tumor.py

@@ -1,5 +1,6 @@
 """`eds.solid_tumor` pipeline"""
 
+from collections import defaultdict
 from typing import Any, Dict, List, Optional, Union
 
 from spacy.tokens import Doc, Span
@@ -108,10 +109,10 @@ def __init__(
             nlp=nlp,
             name=name,
             patterns=patterns,
-            detailed_status_mapping={
-                1: "LOCALIZED",
-                2: "METASTASIS",
-            },
+            detailed_status_mapping=defaultdict(
+                lambda: "LOCALIZED",
+                {2: "METASTASIS"},
+            ),
             label=label,
             span_setter=span_setter,
         )

tests/pipelines/ner/disorders/alcohol.py

@@ -9,6 +9,7 @@
         True,
         True,
         True,
+        True,
     ],
     detailled_status=[
         None,
@@ -20,28 +21,31 @@
         None,
         "ABSTINENCE",
         None,
+        None,
     ],
     negation=[
         None,
         None,
         None,
         None,
         None,
-        None,
+        False,
         True,
         None,
         True,
+        False,
     ],
     assign=None,
     texts=[
         "Patient alcoolique.",
         "OH chronique.",
         "Prise d'alcool occasionnelle",
         "Application d'un pansement alcoolisé",
-        "Alcoolisme sevré",
-        "Alcoolisme non sevré",
+        "Présence d'un alcoolisme sevré",
+        "Présence d'un alcoolisme non sevré",
         "Alcool: 0",
         "Le patient est en cours de sevrage éthylotabagique",
         "Patient alcoolique: non.",
+        "On a un alcoolique non sevré depuis 10 ans.",
     ],
 )

tests/pipelines/ner/disorders/test_all.py

@@ -99,7 +99,10 @@ def check(self):
             for ent in ents:
                 assert ent.label_ == self.disorder
                 if negation is not None:
-                    assert ent._.negation == negation
+                    if negation:
+                        assert ent._.negation == negation
+                    else:
+                        assert ent._.negation is None
 
             if not ents:
                 continue

tests/pipelines/ner/disorders/tobacco.py

@@ -9,6 +9,7 @@
         True,
         True,
         True,
+        True,
     ],
     detailled_status=[
         None,
@@ -20,6 +21,7 @@
         "ABSTINENCE",
         None,
         None,
+        None,
     ],
     negation=[
         None,
@@ -31,17 +33,19 @@
         None,
         True,
         True,
+        False,
     ],
     assign=[{"PA": 15}] + 8 * [None],
     texts=[
         "Tabagisme évalué à 15 PA",
         "Patient tabagique",
         "Tabagisme festif",
         "On a un tabagisme ancien",
-        "Tabac: 0",
-        "Tabagisme passif",
-        "Tabac: sevré depuis 5 ans",
+        "Pour le tabac: 0",
+        "Notion de tabagisme passif",
+        "Concernant le tabac: sevré depuis 5 ans",
         "Le patient ne fume aucun truc.",
         "Le patient fume 0 PA.",
+        "On a un tabagique non sevré depuis 10 ans.",
     ],
 )