Extract Anatomical Entities from Oncology Texts

Description

This model extracts anatomical entities using an unspecific label.

Definitions of Predicted Entities:

  • Anatomical_Site: Relevant anatomical terms mentioned in text.
  • Direction: Directional and laterality terms, such as “left”, “right”, “bilateral”, “upper” and “lower”.

Predicted Entities

Anatomical_Site, Direction

Live Demo Open in Colab Copy S3 URI

How to use

document_assembler = DocumentAssembler()\
    .setInputCol("text")\
    .setOutputCol("document")

sentence_detector = SentenceDetectorDLModel.pretrained("sentence_detector_dl_healthcare","en","clinical/models")\
    .setInputCols(["document"])\
    .setOutputCol("sentence")

tokenizer = Tokenizer() \
    .setInputCols(["sentence"]) \
    .setOutputCol("token")

word_embeddings = WordEmbeddingsModel().pretrained("embeddings_clinical", "en", "clinical/models")\
    .setInputCols(["sentence", "token"]) \
    .setOutputCol("embeddings")                

ner = MedicalNerModel.pretrained("ner_oncology_anatomy_general", "en", "clinical/models") \
    .setInputCols(["sentence", "token", "embeddings"]) \
    .setOutputCol("ner")

ner_converter = NerConverter() \
    .setInputCols(["sentence", "token", "ner"]) \
    .setOutputCol("ner_chunk")
pipeline = Pipeline(stages=[document_assembler,
                            sentence_detector,
                            tokenizer,
                            word_embeddings,
                            ner,
                            ner_converter])

data = spark.createDataFrame([["The patient presented a mass in her left breast, and a possible metastasis in her lungs and in her liver."]]).toDF("text")

result = pipeline.fit(data).transform(data)
val document_assembler = new DocumentAssembler()
    .setInputCol("text")
    .setOutputCol("document")
    
val sentence_detector = SentenceDetectorDLModel.pretrained("sentence_detector_dl_healthcare","en","clinical/models")
    .setInputCols("document")
    .setOutputCol("sentence")
    
val tokenizer = new Tokenizer()
    .setInputCols("sentence")
    .setOutputCol("token")
    
val word_embeddings = WordEmbeddingsModel().pretrained("embeddings_clinical", "en", "clinical/models")
    .setInputCols(Array("sentence", "token"))
    .setOutputCol("embeddings")                
    
val ner = MedicalNerModel.pretrained("ner_oncology_anatomy_general", "en", "clinical/models")
    .setInputCols(Array("sentence", "token", "embeddings"))
    .setOutputCol("ner")
    
val ner_converter = new NerConverter()
    .setInputCols(Array("sentence", "token", "ner"))
    .setOutputCol("ner_chunk")

        
val pipeline = new Pipeline().setStages(Array(document_assembler,
                            sentence_detector,
                            tokenizer,
                            word_embeddings,
                            ner,
                            ner_converter))    

val data = Seq("The patient presented a mass in her left breast, and a possible metastasis in her lungs and in her liver.").toDS.toDF("text")

val result = pipeline.fit(data).transform(data)
import nlu
nlu.load("en.med_ner.anatomy_general").predict("""The patient presented a mass in her left breast, and a possible metastasis in her lungs and in her liver.""")

Results

| chunk   | ner_label       |
|:--------|:----------------|
| left    | Direction       |
| breast  | Anatomical_Site |
| lungs   | Anatomical_Site |
| liver   | Anatomical_Site |

Model Information

Model Name: ner_oncology_anatomy_general
Compatibility: Healthcare NLP 4.2.2+
License: Licensed
Edition: Official
Input Labels: [sentence, token, embeddings]
Output Labels: [ner]
Language: en
Size: 34.3 MB

References

In-house annotated oncology case reports.

Benchmarking

          label   tp  fp  fn  total  precision  recall   f1
Anatomical_Site 2946 549 638   3584       0.84    0.82 0.83
      Direction  864 209 120    984       0.81    0.88 0.84
      macro_avg 3810 758 758   4568       0.82    0.85 0.84
      micro_avg 3810 758 758   4568       0.83    0.83 0.83