Description
This pretrained model maps ICD10CM codes to corresponding UMLS codes under the Unified Medical Language System (UMLS).
Predicted Entities
umls_code
How to use
document_assembler = DocumentAssembler()\
.setInputCol('text')\
.setOutputCol('document')
chunk_assembler = Doc2Chunk()\
.setInputCols(['document'])\
.setOutputCol('ner_chunk')
mapperModel = ChunkMapperModel.pretrained("icd10cm_umls_mapper", "en", "clinical/models")\
.setInputCols(["ner_chunk"])\
.setOutputCol("mappings")\
.setRels(["umls_code"])
mapper_pipeline = Pipeline(stages=[
document_assembler,
chunk_assembler,
mapperModel
])
data = spark.createDataFrame([["A01.2"], ["F10.220"]]).toDF("text")
result = mapper_pipeline.fit(data).transform(data)
document_assembler = nlp.DocumentAssembler()\
.setInputCol('text')\
.setOutputCol('document')
chunk_assembler = medical.Doc2Chunk()\
.setInputCols(['document'])\
.setOutputCol('ner_chunk')
mapperModel = medical.ChunkMapperModel.pretrained("icd10cm_umls_mapper", "en", "clinical/models")\
.setInputCols(["ner_chunk"])\
.setOutputCol("mappings")\
.setRels(["umls_code"])
mapper_pipeline = Pipeline(stages=[
document_assembler,
chunk_assembler,
mapperModel
])
data = spark.createDataFrame([["A01.2"], ["F10.220"]]).toDF("text")
result = mapper_pipeline.fit(data).transform(data)
val document_assembler = new DocumentAssembler()
.setInputCol("text")
.setOutputCol("document")
val chunk_assembler = new Doc2Chunk()
.setInputCols("document")
.setOutputCol("ner_chunk")
val chunkerMapper = ChunkMapperModel
.pretrained("icd10cm_umls_mapper", "en", "clinical/models")
.setInputCols(Array("ner_chunk"))
.setOutputCol("mappings")
.setRels(Array("umls_code"))
val mapper_pipeline = new Pipeline().setStages(Array(
document_assembler,
chunk_assembler,
chunkerMapper))
val data = Seq("A01.2", "F10.220").toDF("text")
val result = mapper_pipeline.fit(data).transform(data)
Results
+------------+---------+---------+
|icd10cm_code|umls_code| relation|
+------------+---------+---------+
| A01.2| C0343376|umls_code|
| F10.220| C2874385|umls_code|
+------------+---------+---------+
Model Information
| Model Name: | icd10cm_umls_mapper |
| Compatibility: | Healthcare NLP 6.0.2+ |
| License: | Licensed |
| Edition: | Official |
| Input Labels: | [ner_chunk] |
| Output Labels: | [mappings] |
| Language: | en |
| Size: | 1.4 MB |
References
Trained on concepts from ICD10CM for the 2025AA release of the Unified Medical Language System® (UMLS) Knowledge Sources: https://www.nlm.nih.gov/research/umls/index.html