Relation extraction between body parts and procedures

Description

Relation extraction between body parts entities like ‘Internal_organ_or_component’, ’External_body_part_or_region’ etc. and procedure and test entities.

Predicted Entities

1 : body part and test/procedure are related to each other. 0 : body part and test/procedure are not related to each other.

Open in Colab Download

How to use

...
words_embedder = WordEmbeddingsModel() \
    .pretrained("embeddings_clinical", "en", "clinical/models") \
    .setInputCols(["sentences", "tokens"]) \
    .setOutputCol("embeddings")
ner_tagger = NerDLModel() \
    .pretrained("jsl_ner_wip_greedy_clinical", "en", "clinical/models") \
    .setInputCols(["sentences", "tokens", "embeddings"]) \
    .setOutputCol("ner_tags")
ner_converter = NerConverter() \
    .setInputCols(["sentences", "tokens", "ner_tags"]) \
    .setOutputCol("ner_chunks")
dependency_parser = DependencyParserModel() \
    .pretrained("dependency_conllu", "en") \
    .setInputCols(["sentences", "pos_tags", "tokens"]) \
    .setOutputCol("dependencies")

# Set a filter on pairs of named entities which will be treated as relation candidates
re_ner_chunk_filter = RENerChunksFilter() \
    .setInputCols(["ner_chunks", "dependencies"])\
    .setMaxSyntacticDistance(10)\
    .setOutputCol("re_ner_chunks").setRelationPairs(["external_body_part_or_region-test"])

# The dataset this model is trained to is sentence-wise. 
# This model can also be trained on document-level relations - in which case, while predicting, use "document" instead of "sentence" as input.
re_model = RelationExtractionDLModel()\
    .pretrained('redl_bodypart_procedure_test_biobert', 'en', "clinical/models") \
    .setPredictionThreshold(0.5)\
    .setInputCols(["re_ner_chunks", "sentences"]) \
    .setOutputCol("relations")

pipeline = Pipeline(stages=[documenter, sentencer, tokenizer, pos_tagger, words_embedder, ner_tagger, ner_converter, dependency_parser, re_ner_chunk_filter, re_model])

text ="TECHNIQUE IN DETAIL: After informed consent was obtained from the patient and his mother, the chest was scanned with portable ultrasound."
p_model = pipeline.fit(spark.createDataFrame([[text]]).toDF("text"))
result = p_model.transform(data)
...
val words_embedder = WordEmbeddingsModel()
    .pretrained("embeddings_clinical", "en", "clinical/models")
    .setInputCols(Array("sentences", "tokens"))
    .setOutputCol("embeddings")
val ner_tagger = NerDLModel()
    .pretrained("ner_clinical", "en", "clinical/models")
    .setInputCols(Array("sentences", "tokens", "embeddings"))
    .setOutputCol("ner_tags")
val ner_converter = NerConverter()
    .setInputCols(Array("sentences", "tokens", "ner_tags"))
    .setOutputCol("ner_chunks")
val dependency_parser = DependencyParserModel()
    .pretrained("dependency_conllu", "en")
    .setInputCols(Array("sentences", "pos_tags", "tokens"))
    .setOutputCol("dependencies")

// Set a filter on pairs of named entities which will be treated as relation candidates
val re_ner_chunk_filter = RENerChunksFilter()
    .setInputCols(Array("ner_chunks", "dependencies"))
    .setMaxSyntacticDistance(10)
    .setOutputCol("re_ner_chunks").setRelationPairs(Array("external_body_part_or_region-test"))

// The dataset this model is trained to is sentence-wise. 
// This model can also be trained on document-level relations - in which case, while predicting, use "document" instead of "sentence" as input.
val re_model = RelationExtractionDLModel()
    .pretrained("redl_bodypart_procedure_test_biobert", "en", "clinical/models")
    .setPredictionThreshold(0.5)
    .setInputCols(Array("re_ner_chunks", "sentences"))
    .setOutputCol("relations")
val pipeline = new Pipeline().setStages(Array(documenter, sentencer, tokenizer, pos_tagger, words_embedder, ner_tagger, ner_converter, dependency_parser, re_ner_chunk_filter, re_model))

val result = pipeline.fit(Seq.empty["TECHNIQUE IN DETAIL: After informed consent was obtained from the patient and his mother, the chest was scanned with portable ultrasound."].toDS.toDF("text")).transform(data)

Results

|    |   relation | entity1                      | chunk1   | entity2   | chunk2              |   confidence |
|---:|-----------:|:-----------------------------|:---------|:----------|:--------------------|-------------:|
|  0 |          1 | External_body_part_or_region | chest    | Test      | portable ultrasound |      0.99953 |

Model Information

Model Name: redl_bodypart_procedure_test_biobert
Compatibility: Spark NLP 2.7.3+
License: Licensed
Edition: Official
Language: en

Data Source

Trained on a custom internal dataset.

Benchmarking

Relation           Recall Precision        F1   Support
0                   0.338     0.472     0.394       325
1                   0.904     0.843     0.872      1275
Avg.                0.621     0.657     0.633