Description
This model maps UMLS codes to ICD-10-PCS codes. It is trained on the 2026AA release of the Unified Medical Language System (UMLS) dataset.
Live Demo Open in Colab Copy S3 URI
How to use
document_assembler = DocumentAssembler()\
.setInputCol("text")\
.setOutputCol("doc")
doc2chunk = Doc2Chunk()\
.setInputCols(["doc"])\
.setOutputCol("ner_chunk")
mapper = ChunkMapperModel.pretrained("umls_icd10pcs_mapper","en","clinical/models")\
.setInputCols(["ner_chunk"])\
.setOutputCol("mappings")
pipeline = Pipeline(stages=[document_assembler, doc2chunk, mapper])
data = spark.createDataFrame([["C0011931"],["C0015618"],["C0029565"]]).toDF("text")
result = pipeline.fit(data).transform(data)
document_assembler = nlp.DocumentAssembler()\
.setInputCol("text")\
.setOutputCol("doc")
doc2chunk = nlp.Doc2Chunk()\
.setInputCols(["doc"])\
.setOutputCol("ner_chunk")
mapper = medical.ChunkMapperModel.pretrained("umls_icd10pcs_mapper","en","clinical/models")\
.setInputCols(["ner_chunk"])\
.setOutputCol("mappings")
pipeline = nlp.Pipeline(stages=[document_assembler, doc2chunk, mapper])
data = spark.createDataFrame([["C0011931"],["C0015618"],["C0029565"]]).toDF("text")
result = pipeline.fit(data).transform(data)
val documentAssembler = new DocumentAssembler()
.setInputCol("text")
.setOutputCol("doc")
val doc2chunk = new Doc2Chunk()
.setInputCols(Array("doc"))
.setOutputCol("ner_chunk")
val mapper = ChunkMapperModel.pretrained("umls_icd10pcs_mapper","en","clinical/models")
.setInputCols(Array("ner_chunk"))
.setOutputCol("mappings")
val pipeline = new Pipeline().setStages(Array(documentAssembler, doc2chunk, mapper))
val data = Seq("C0011931","C0015618","C0029565").toDF("text")
val result = pipeline.fit(data).transform(data)
Results
| umls_code | icd10pcs_code |
|:------------|:----------------|
| C0011931 | BH4CZZZ |
| C0015618 | GZ72ZZZ |
| C0029565 | GZ63ZZZ |
Model Information
| Model Name: | umls_icd10pcs_mapper |
| Compatibility: | Healthcare NLP 6.4.0+ |
| License: | Licensed |
| Edition: | Official |
| Input Labels: | [ner_chunk] |
| Output Labels: | [mappings] |
| Language: | en |
| Size: | 2.7 MB |