Mapping ICD-10-CM Codes with Corresponding Billable and Hierarchical Condition Category (HCC) Scores

Description

This pretrained model maps ICD-10-CM codes with their corresponding billable and HCC scores. If there is no HCC score for the corresponding ICD-10-CM code, result will be returned as 0. Trained on the ICD-10-CM 20260401 (FY2026, effective April 1, 2026) code set.

Live Demo Open in Colab Copy S3 URI

How to use


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

icd10cm_billable_hcc_mapper = DocMapperModel.pretrained("icd10cm_billable_hcc_mapper", "en", "clinical/models")\
    .setInputCols(["document"])\
    .setOutputCol("mappings")\
    .setRels(['billable', 'hcc_score'])\
    .setLowerCase(True)\
    .setMultivaluesRelations(True)

pipeline = Pipeline(stages=[document_assembler, icd10cm_billable_hcc_mapper])
data = spark.createDataFrame([["A00.0"]]).toDF("text")
result = pipeline.fit(data).transform(data)


document_assembler = nlp.DocumentAssembler()\
    .setInputCol("text")\
    .setOutputCol("document")

icd10cm_billable_hcc_mapper = medical.DocMapperModel.pretrained("icd10cm_billable_hcc_mapper", "en", "clinical/models")\
    .setInputCols(["document"])\
    .setOutputCol("mappings")\
    .setRels(['billable', 'hcc_score'])\
    .setLowerCase(True)\
    .setMultivaluesRelations(True)

pipeline = nlp.Pipeline(stages=[document_assembler, icd10cm_billable_hcc_mapper])
data = spark.createDataFrame([["A00.0"]]).toDF("text")
result = pipeline.fit(data).transform(data)


val documentAssembler = new DocumentAssembler()
    .setInputCol("text")
    .setOutputCol("document")

val icd10cm_billable_hcc_mapper = DocMapperModel.pretrained("icd10cm_billable_hcc_mapper", "en", "clinical/models")
    .setInputCols(Array("document"))
    .setOutputCol("mappings")
    .setRels(Array("billable", "hcc_score"))
    .setLowerCase(true)
    .setMultivaluesRelations(true)

val pipeline = new Pipeline().setStages(Array(documentAssembler, icd10cm_billable_hcc_mapper))
val data = Seq("A00.0").toDF("text")
val result = pipeline.fit(data).transform(data)

Results

| code   |   billable |   hcc_score |
|:-------|-----------:|------------:|
| A00.0  |          1 |           0 |
| E11.9  |          1 |          38 |
| I10    |          1 |           0 |
| J44.9  |          1 |         280 |
| K35.20 |          0 |           0 |

Model Information

Model Name: icd10cm_billable_hcc_mapper
Compatibility: Healthcare NLP 6.4.0+
License: Licensed
Edition: Official
Input Labels: [document]
Output Labels: [mappings]
Language: en
Size: 1.1 MB