Edit on GitHub
  • John Snow Labs
  • May 01, 2025

Human Phenotypes Text Matcher

  • en
  • licensed
  • clinical
  • hpo

Description

This model is a text matcher designed to automatically extract mentions of phenotypic abnormalities associated with human diseases from clinical or biomedical text.

Predicted Entities

Copy S3 URI

How to use

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

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

stopwords_cleaner = StopWordsCleaner()\
    .setInputCols("token")\
    .setOutputCol("cleanTokens")\
    .setCaseSensitive(False)

token_assembler = TokenAssembler()\
    .setInputCols(['document',"cleanTokens"])\
    .setOutputCol("cleanTokens_newDoc")

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

tokenizer_2 = Tokenizer()\
    .setInputCols(["sentence"])\
    .setOutputCol("clean_tokens")

entityExtractor = TextMatcherInternalModel().pretrained("hpo_matcher","en","clinical/models")\
    .setInputCols(["sentence", "clean_tokens"])\
    .setOutputCol("hpo_term")\
    .setCaseSensitive(False)\
    .setMergeOverlapping(False)

matcher_pipeline = Pipeline().setStages([
                  documentAssembler,                  
                  tokenizer,
                  stopwords_cleaner,
                  token_assembler,
                  sentenceDetector,
                  tokenizer_2,
                  entityExtractor])

text =  ''' APNEA: Presumed apnea of prematurity since < 34 wks gestation at birth.
GENETICS: Holds thumbs in palms, findings on Hemolytic Uremic Syndrome, history of meconium plugs.
HYPERBILIRUBINEMIA: At risk for hyperbilirubinemia d/t prematurity. 
1/25-1/30: Received Amp/Gent while undergoing sepsis evaluation. '''

data = spark.createDataFrame([[text]]).toDF("text")
 
matcher_model = matcher_pipeline.fit(data).transform(data)

documentAssembler = nlp.DocumentAssembler()\
    .setInputCol("text")\
    .setOutputCol("document")

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

stopwords_cleaner = nlp.StopWordsCleaner()\
    .setInputCols("token")\
    .setOutputCol("cleanTokens")\
    .setCaseSensitive(False)

token_assembler = nlp.TokenAssembler()\
    .setInputCols(['document',"cleanTokens"])\
    .setOutputCol("cleanTokens_newDoc")

sentenceDetector = nlp.SentenceDetectorDLModel.pretrained("sentence_detector_dl_healthcare","en","clinical/models") \
    .setInputCols(["cleanTokens_newDoc"]) \
    .setOutputCol("sentence") 

tokenizer_2 = nlp.Tokenizer()\
    .setInputCols(["sentence"])\
    .setOutputCol("clean_tokens")

entityExtractor = medical.TextMatcherInternalModel().pretrained("hpo_matcher","en","clinical/models")\
    .setInputCols(["sentence", "clean_tokens"])\
    .setOutputCol("hpo_term")\
    .setCaseSensitive(False)\
    .setMergeOverlapping(False)

matcher_pipeline = nlp.Pipeline().setStages([
                  documentAssembler,                  
                  tokenizer,
                  stopwords_cleaner,
                  token_assembler,
                  sentenceDetector,
                  tokenizer_2,
                  entityExtractor])

text =  ''' APNEA: Presumed apnea of prematurity since < 34 wks gestation at birth.
GENETICS: Holds thumbs in palms, findings on Hemolytic Uremic Syndrome, history of meconium plugs.
HYPERBILIRUBINEMIA: At risk for hyperbilirubinemia d/t prematurity. 
1/25-1/30: Received Amp/Gent while undergoing sepsis evaluation. '''

data = spark.createDataFrame([[text]]).toDF("text")
 
matcher_model = matcher_pipeline.fit(data).transform(data)

val documentAssembler = new DocumentAssembler()
  .setInputCol("text")
  .setOutputCol("document")

val tokenizer = new Tokenizer()
  .setInputCols("document")
  .setOutputCol("token")

val stopwordsCleaner = new StopWordsCleaner()
  .setInputCols("token")
  .setOutputCol("cleanTokens")
  .setCaseSensitive(false)

val tokenAssembler = new TokenAssembler()
  .setInputCols(Array("document", "cleanTokens"))
  .setOutputCol("cleanTokens_newDoc")

val sentenceDetector = SentenceDetectorDLModel
  .pretrained("sentence_detector_dl_healthcare", "en", "clinical/models")
  .setInputCols(Array("cleanTokens_newDoc"))
  .setOutputCol("sentence")

val tokenizer2 = new Tokenizer()
  .setInputCols("sentence")
  .setOutputCol("clean_tokens")

val entityExtractor = TextMatcherInternalModel
  .pretrained("hpo_matcher", "en", "clinical/models")
  .setInputCols(Array("sentence", "clean_tokens"))
  .setOutputCol("hpo_term")
  .setCaseSensitive(false)
  .setMergeOverlapping(false)

val pipeline = new Pipeline().setStages(Array(
  documentAssembler,
  tokenizer,
  stopwordsCleaner,
  tokenAssembler,
  sentenceDetector,
  tokenizer2,
  entityExtractor
))

val data = Seq("""APNEA: Presumed apnea of prematurity since < 34 wks gestation at birth.
GENETICS: Holds thumbs in palms, findings on Hemolytic Uremic Syndrome, history of meconium plugs.
HYPERBILIRUBINEMIA: At risk for hyperbilirubinemia d/t prematurity. 
1/25-1/30: Received Amp/Gent while undergoing sepsis evaluation.""").toDF("text")

val matcher = pipeline.fit(data).transform(data)

Results

+-------------------------+-----+---+-----+
|                    chunk|begin|end|label|
+-------------------------+-----+---+-----+
|                    APNEA|    0|  4|  HPO|
|                    apnea|   16| 20|  HPO|
|Hemolytic Uremic Syndrome|  105|129|  HPO|
|       HYPERBILIRUBINEMIA|  156|173|  HPO|
|       hyperbilirubinemia|  181|198|  HPO|
|                   sepsis|  257|262|  HPO|
+-------------------------+-----+---+-----+

Model Information

Model Name: hpo_matcher
Compatibility: Healthcare NLP 6.0.0+
License: Licensed
Edition: Official
Input Labels: [document, token]
Output Labels: [hpo_term]
Language: en
Size: 2.1 MB
Case sensitive: false