Description
This model can be used to detect normalized mentions of genes (go) and human phenotypes (hp) in medical text.
Predicted Entities
GO, HP
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_clinical", "en", "clinical/models")\
.setInputCols(["sentence", "token"])\
.setOutputCol("embeddings")
clinical_ner = NerDLModel.pretrained("ner_human_phenotype_go_clinical", "en", "clinical/models") \
.setInputCols(["sentence", "token", "embeddings"]) \
.setOutputCol("ner")
...
nlp_pipeline = Pipeline(stages=[document_assembler, sentence_detector, tokenizer, word_embeddings, clinical_ner, ner_converter])
light_pipeline = LightPipeline(nlp_pipeline.fit(spark.createDataFrame([['']]).toDF("text")))
annotations = light_pipeline.fullAnnotate("Another disease that shares two of the tumor components of CT, namely GIST and tricarboxylic acid cycle is the Carney-Stratakis syndrome (CSS) or dyad.")
...
val word_embeddings = WordEmbeddingsModel.pretrained("embeddings_clinical", "en", "clinical/models")
.setInputCols(Array("sentence", "token"))
.setOutputCol("embeddings")
val ner = NerDLModel.pretrained("ner_human_phenotype_go_clinical", "en", "clinical/models")
.setInputCols("sentence", "token", "embeddings")
.setOutputCol("ner")
...
val pipeline = new Pipeline().setStages(Array(document_assembler, sentence_detector, tokenizer, word_embeddings, ner, ner_converter))
val data = Seq("Another disease that shares two of the tumor components of CT, namely GIST and tricarboxylic acid cycle is the Carney-Stratakis syndrome (CSS) or dyad.").toDF("text")
val result = pipeline.fit(data).transform(data)
import nlu
nlu.load("en.med_ner.human_phenotype.go_clinical").predict("""Another disease that shares two of the tumor components of CT, namely GIST and tricarboxylic acid cycle is the Carney-Stratakis syndrome (CSS) or dyad.""")
Results
+----+--------------------------+---------+-------+----------+
| | chunk | begin | end | entity |
+====+==========================+=========+=======+==========+
| 0 | tumor | 39 | 43 | HP |
+----+--------------------------+---------+-------+----------+
| 1 | tricarboxylic acid cycle | 79 | 102 | GO |
+----+--------------------------+---------+-------+----------+
Model Information
| Model Name: | ner_human_phenotype_go_clinical |
| Type: | ner |
| Compatibility: | Healthcare NLP 2.6.0 + |
| Edition: | Official |
| License: | Licensed |
| Input Labels: | [sentence, token, embeddings] |
| Output Labels: | [ner] |
| Language: | [en] |
| Case sensitive: | false |
| Dependencies: | embeddings_clinical |
Benchmarking
| | label | tp | fp | fn | prec | rec | f1 |
|---:|--------------:|------:|-----:|-----:|---------:|---------:|---------:|
| 0 | B-GO | 1530 | 129 | 57 | 0.922242 | 0.964083 | 0.942699 |
| 1 | B-HP | 950 | 133 | 130 | 0.877193 | 0.87963 | 0.87841 |
| 2 | I-HP | 253 | 46 | 68 | 0.846154 | 0.788162 | 0.816129 |
| 3 | I-GO | 4550 | 344 | 154 | 0.92971 | 0.967262 | 0.948114 |
| 4 | Macro-average | 7283 | 652 | 409 | 0.893825 | 0.899784 | 0.896795 |
| 5 | Micro-average | 7283 | 652 | 409 | 0.917832 | 0.946828 | 0.932105 |