Mapping Entities with Corresponding RxNorm Codes

Description

This pretrained model maps entities with their corresponding RxNorm codes.

Predicted Entities

RxNorm Codes

Open in Colab Copy S3 URI

How to use

document_assembler = DocumentAssembler()\
.setInputCol('text')\
.setOutputCol('document')

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

tokenizer = Tokenizer()\
    .setInputCols("sentence")\
    .setOutputCol("token")

word_embeddings = WordEmbeddingsModel\
    .pretrained("embeddings_clinical", "en", "clinical/models")\
    .setInputCols(["sentence", "token"])\
    .setOutputCol("embeddings")

posology_ner_model = MedicalNerModel\
    .pretrained("ner_posology_greedy", "en", "clinical/models")\
    .setInputCols(["sentence", "token", "embeddings"])\
    .setOutputCol("posology_ner")

posology_ner_converter = NerConverterInternal()\
    .setInputCols("sentence", "token", "posology_ner")\
    .setOutputCol("ner_chunk")

chunkerMapper = ChunkMapperModel\
    .pretrained("rxnorm_mapper", "en", "clinical/models")\
    .setInputCols(["ner_chunk"])\
    .setOutputCol("mappings")\
    .setRels(["rxnorm_code"])

mapper_pipeline = Pipeline().setStages([
document_assembler,
sentence_detector,
tokenizer, 
word_embeddings,
posology_ner_model, 
posology_ner_converter, 
chunkerMapper])


test_data = spark.createDataFrame([["The patient was given Zyrtec 10 MG, Adapin 10 MG Oral Capsule, Septi-Soothe 0.5 Topical Spray"]]).toDF("text")

mapper_model = mapper_pipeline.fit(test_data)

result= mapper_model.transform(test_data)

Results

+------------------------------+-----------+
|ner_chunk                     |rxnorm_code|
+------------------------------+-----------+
|Zyrtec 10 MG                  |1011483    |
|Adapin 10 MG Oral Capsule     |1000050    |
|Septi-Soothe 0.5 Topical Spray|1000046    |
+------------------------------+-----------+

Model Information

Model Name: rxnorm_mapper
Compatibility: Healthcare NLP 5.3.3+
License: Licensed
Edition: Official
Input Labels: [ner_chunk]
Output Labels: [mappings]
Language: en
Size: 10.5 MB