Description
Pretrained named entity recognition deep learning model for healthcare. Includes Problem, Test and Treatment entities. The SparkNLP deep learning model (NerDL) is inspired by a former state of the art model for NER: Chiu & Nicols, Named Entity Recognition with Bidirectional LSTM-CNN.
Predicted Entities
PROBLEM, TEST, TREATMENT.
Live Demo Open in Colab Copy S3 URI
How to use
Use as part of an nlp pipeline with the following stages: DocumentAssembler, SentenceDetector, Tokenizer, WordEmbeddingsModel, NerDLModel. Add the NerConverter to the end of the pipeline to convert entity tokens into full entity chunks.
...
word_embeddings = WordEmbeddingsModel.pretrained("embeddings_healthcare_100d", "en", "clinical/models")\
.setInputCols(["sentence", "token"])\
.setOutputCol("embeddings")
clinical_ner = NerDLModel.pretrained("ner_healthcare", "en", "clinical/models") \
.setInputCols(["sentence", "token", "embeddings"]) \
.setOutputCol("ner")
...
nlpPipeline = Pipeline(stages=[document_assembler, sentence_detector, tokenizer, word_embeddings, clinical_ner, ner_converter])
model = nlpPipeline.fit(spark.createDataFrame([["A 28-year-old female with a history of gestational diabetes mellitus diagnosed eight years prior to presentation and subsequent type two diabetes mellitus ( T2DM ), one prior episode of HTG-induced pancreatitis three years prior to presentation , associated with an acute hepatitis , and obesity with a body mass index ( BMI ) of 33.5 kg/m2 , presented with a one-week history of polyuria , polydipsia , poor appetite , and vomiting . Two weeks prior to presentation , she was treated with a five-day course of amoxicillin for a respiratory tract infection . She was on metformin , glipizide , and dapagliflozin for T2DM and atorvastatin and gemfibrozil for HTG ."]]).toDF("text"))
results = model.transform(data)
...
val word_embeddings = WordEmbeddingsModel.pretrained("embeddings_healthcare_100d", "en", "clinical/models")
.setInputCols(Array("sentence", "token"))
.setOutputCol("embeddings")
val ner = NerDLModel.pretrained("ner_healthcare", "en", "clinical/models")
.setInputCols(Array("sentence", "token", "embeddings"))
.setOutputCol("ner")
...
val pipeline = new Pipeline().setStages(Array(document_assembler, sentence_detector, tokenizer, word_embeddings, ner, ner_converter))
val data = Seq(A 28-year-old female with a history of gestational diabetes mellitus diagnosed eight years prior to presentation and subsequent type two diabetes mellitus ( T2DM ), one prior episode of HTG-induced pancreatitis three years prior to presentation , associated with an acute hepatitis , and obesity with a body mass index ( BMI ) of 33.5 kg/m2 , presented with a one-week history of polyuria , polydipsia , poor appetite , and vomiting . Two weeks prior to presentation , she was treated with a five-day course of amoxicillin for a respiratory tract infection . She was on metformin , glipizide , and dapagliflozin for T2DM and atorvastatin and gemfibrozil for HTG .).toDF("text")
val result = pipeline.fit(data).transform(data)
Results
The output is a dataframe with a sentence per row and a "ner" column containing all of the entity labels in the sentence, entity character indices, and other metadata. To get only the tokens and entity labels, without the metadata, select "token.result" and "ner.result" from your output dataframe or add the "Finisher" to the end of your pipeline.
| | chunk | ner_label |
|---|-------------------------------|-----------|
| 0 | a respiratory tract infection | PROBLEM |
| 1 | metformin | TREATMENT |
| 2 | glipizide | TREATMENT |
| 3 | dapagliflozin | TREATMENT |
| 4 | T2DM | PROBLEM |
| 5 | atorvastatin | TREATMENT |
| 6 | gemfibrozil | TREATMENT |
Model Information
| Model Name: | ner_healthcare_en_2.4.4_2.4 |
| Type: | ner |
| Compatibility: | Spark NLP 2.4.4 |
| Edition: | Official |
| License: | Licensed |
| Input Labels: | [sentence,token, embeddings] |
| Output Labels: | [ner] |
| Language: | [en] |
| Case sensitive: | false |
Data Source
Trained with augmented version of i2b2 dataset with ‘embeddings_healthcare_100d’. https://portal.dbmi.hms.harvard.edu/projects/n2c2-nlp/
Benchmarking
| | label | tp | fp | fn | prec | rec | f1 |
|---:|:--------------|------:|------:|------:|---------:|---------:|---------:|
| 0 | I-TREATMENT | 6625 | 1187 | 1329 | 0.848054 | 0.832914 | 0.840416 |
| 1 | I-PROBLEM | 15142 | 1976 | 2542 | 0.884566 | 0.856254 | 0.87018 |
| 2 | B-PROBLEM | 11005 | 1065 | 1587 | 0.911765 | 0.873968 | 0.892466 |
| 3 | I-TEST | 6748 | 923 | 1264 | 0.879677 | 0.842237 | 0.86055 |
| 4 | B-TEST | 8196 | 942 | 1029 | 0.896914 | 0.888455 | 0.892665 |
| 5 | B-TREATMENT | 8271 | 1265 | 1073 | 0.867345 | 0.885167 | 0.876165 |
| 6 | Macro-average | 55987 | 7358 | 8824 | 0.881387 | 0.863166 | 0.872181 |
| 7 | Micro-average | 55987 | 7358 | 8824 | 0.883842 | 0.86385 | 0.873732 |