Description
Relate scale items and their measurements according to NIHSS guidelines.
Predicted Entities
Has_Value : Measurement is related to the entity, 0 : Measurement is not related to the entity
Live Demo Open in Colab Copy S3 URI
How to use
documenter = DocumentAssembler()\
.setInputCol("text")\
.setOutputCol("document")
sentencer = SentenceDetector()\
.setInputCols(["document"])\
.setOutputCol("sentences")
tokenizer = sparknlp.annotators.Tokenizer()\
.setInputCols(["sentences"])\
.setOutputCol("tokens")
pos_tagger = PerceptronModel().pretrained("pos_clinical", "en", "clinical/models") \
.setInputCols(["sentences", "tokens"])\
.setOutputCol("pos_tags")
words_embedder = WordEmbeddingsModel().pretrained("embeddings_clinical", "en", "clinical/models") \
.setInputCols(["sentences", "tokens"]) \
.setOutputCol("embeddings")
ner_tagger = MedicalNerModel.pretrained("ner_nihss", "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")
re_model = RelationExtractionDLModel().pretrained('redl_nihss_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= "There , her initial NIHSS score was 4 , as recorded by the ED physicians . This included 2 for weakness in her left leg and 2 for what they felt was subtle ataxia in her left arm and leg ."
p_model = pipeline.fit(spark.createDataFrame([[text]]).toDF("text"))
result = p_model.transform(spark.createDataFrame(pd.DataFrame({'text': [text]})))
val documenter = new DocumentAssembler()
.setInputCol("text")
.setOutputCol("document")
val sentencer = new SentenceDetector()
.setInputCols("document")
.setOutputCol("sentences")
val tokenizer = new Tokenizer()
.setInputCols("sentences")
.setOutputCol("tokens")
val pos_tagger = PerceptronModel().pretrained("pos_clinical", "en", "clinical/models")
.setInputCols(Array("sentences", "tokens"))
.setOutputCol("pos_tags")
val words_embedder = WordEmbeddingsModel().pretrained("embeddings_clinical", "en", "clinical/models")
.setInputCols(Array("sentences", "tokens"))
.setOutputCol("embeddings")
val ner_tagger = MedicalNerModel.pretrained("ner_nihss", "en", "clinical/models")
.setInputCols(Array("sentences", "tokens", "embeddings"))
.setOutputCol("ner_tags")
val ner_converter = new 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 = new RENerChunksFilter()
.setInputCols(Array("ner_chunks", "dependencies"))
.setMaxSyntacticDistance(10)
.setOutputCol("re_ner_chunks")
val re_model = RelationExtractionDLModel().pretrained("redl_nihss_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 data = Seq("""There , her initial NIHSS score was 4 , as recorded by the ED physicians . This included 2 for weakness in her left leg and 2 for what they felt was subtle ataxia in her left arm and leg .""").toDS.toDF("text")
val result = pipeline.fit(data).transform(data)
import nlu
nlu.load("en.extract_relation.nihss").predict("""There , her initial NIHSS score was 4 , as recorded by the ED physicians . This included 2 for weakness in her left leg and 2 for what they felt was subtle ataxia in her left arm and leg .""")
Results
+---------+-----------+-------------+-----------+-----------+------------+-------------+-----------+--------------------+----------+
| relation| entity1|entity1_begin|entity1_end| chunk1| entity2|entity2_begin|entity2_end| chunk2|confidence|
+---------+-----------+-------------+-----------+-----------+------------+-------------+-----------+--------------------+----------+
|Has_Value| NIHSS| 20| 30|NIHSS score| Measurement| 36| 36| 4| 0.9998851|
|Has_Value|Measurement| 89| 89| 2| 6a_LeftLeg| 111| 118| left leg| 0.9987311|
| 0|Measurement| 89| 89| 2| Measurement| 124| 124| 2|0.97510725|
| 0|Measurement| 89| 89| 2|7_LimbAtaxia| 156| 185|ataxia in her lef...| 0.999889|
| 0| 6a_LeftLeg| 111| 118| left leg| Measurement| 124| 124| 2|0.99989617|
| 0| 6a_LeftLeg| 111| 118| left leg|7_LimbAtaxia| 156| 185|ataxia in her lef...| 0.9999521|
|Has_Value|Measurement| 124| 124| 2|7_LimbAtaxia| 156| 185|ataxia in her lef...| 0.9896319|
+---------+-----------+-------------+-----------+-----------+------------+-------------+-----------+--------------------+----------+
Model Information
| Model Name: | redl_nihss_biobert |
| Compatibility: | Healthcare NLP 3.3.2+ |
| License: | Licensed |
| Edition: | Official |
| Language: | en |
Data Source
@article{wangnational, title={National Institutes of Health Stroke Scale (NIHSS) Annotations for the MIMIC-III Database}, author={Wang, Jiayang and Huang, Xiaoshuo and Yang, Lin and Li, Jiao} }
Benchmarking
Relation Recall Precision F1 Support
0 0.989 0.976 0.982 611
Has_Value 0.983 0.992 0.988 889
Avg. 0.986 0.984 0.985 -
Weighted-Avg. 0.985 0.985 0.985 -