Pipeline to Mapping UMLS Codes with Their Corresponding SNOMED Codes

Description

This pretrained pipeline is built on the top of umls_snomed_mapper model and maps UMLS codes to corresponding SNOMED codes.

Live Demo Open in Colab Download Copy S3 URI

How to use

from sparknlp.pretrained import PretrainedPipeline

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

sample_text = """ [['acebutolol'], ['aspirin']]"""

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

from johnsnowlabs import nlp, medical

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

sample_text = """ [['acebutolol'], ['aspirin']]"""

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

import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline

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

val sample_text = """ [['acebutolol'], ['aspirin']]"""

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

Results

| chunk      | umls_code | snomed_code |
| :--------- | :-------- | ----------: |
| acebutolol | C3208139  |   767585002 |
| aspirin    | C0732305  |   358427004 |

Model Information

Included Models

• DocumentAssembler
• BertSentenceEmbeddings
• SentenceEntityResolverModel
• Resolution2Chunk
• ChunkMapperModel