Pipeline to Mapping Entities (Drug Substances) with Corresponding UMLS CUI Codes

Description

This pretrained pipeline is built on the top of umls_drug_substance_mapper model and maps entities (Drug Substances) with corresponding UMLS CUI codes.

Live Demo Open in Colab Download Copy S3 URI

How to use

from sparknlp.pretrained import PretrainedPipeline

pipeline = PretrainedPipeline("umls_drug_substance_mapping", "en", "clinical/models")

sample_text = """ The patient was given  metformin, lenvatinib and gallopamil 50 MG Oral Tablet."""

result = pipeline.transform(spark.createDataFrame([[sample_text]]).toDF("text"))

from johnsnowlabs import nlp, medical

pipeline = nlp.PretrainedPipeline("umls_drug_substance_mapping", "en", "clinical/models")

sample_text = """ The patient was given  metformin, lenvatinib and gallopamil 50 MG Oral Tablet."""

result = pipeline.transform(spark.createDataFrame([[sample_text]]).toDF("text"))

import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline

val pipeline = PretrainedPipeline("umls_drug_substance_mapping", "en", "clinical/models")

val sample_text = """ The patient was given  metformin, lenvatinib and gallopamil 50 MG Oral Tablet."""

val result = pipeline.transform(spark.createDataFrame([[sample_text]]).toDF("text"))

Results

| ner_chunk                    | umls_code |
| :--------------------------- | :-------- |
| metformin                    | C0025598  |
| lenvatinib                   | C2986924  |
| gallopamil 50 MG Oral Tablet | C0787234  |

Model Information

Included Models

• DocumentAssembler
• SentenceDetector
• TokenizerModel
• WordEmbeddingsModel
• MedicalNerModel
• NerConverterInternalModel
• ChunkMapperModel