Description
This pretrained pipeline is built on the top of umls_icd10cm_mapper model and maps maps UMLS codes to corresponding ICD10CM codes
How to use
from sparknlp.pretrained import PretrainedPipeline
pipeline = PretrainedPipeline("umls_icd10cm_mapping", "en", "clinical/models")
sample_text = """ [['C0000744'], ['C2875181']]"""
result = pipeline.transform(spark.createDataFrame(sample_text).toDF("text"))
from johnsnowlabs import nlp, medical
pipeline = nlp.PretrainedPipeline("umls_icd10cm_mapping", "en", "clinical/models")
sample_text = """ [['C0000744'], ['C2875181']]"""
result = pipeline.transform(spark.createDataFrame(sample_text).toDF("text"))
import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline
val pipeline = PretrainedPipeline("umls_icd10cm_mapping", "en", "clinical/models")
val sample_text = """ [['C0000744'], ['C2875181']]"""
val result = pipeline.transform(spark.createDataFrame(sample_text).toDF("text"))
Results
| umls_code | icd10cm_code |
| :-------- | :----------- |
| C0000744 | E78.6 |
| C2875181 | G43.81 |
Model Information
| Model Name: | umls_icd10cm_mapping |
| Type: | pipeline |
| Compatibility: | Healthcare NLP 6.3.0+ |
| License: | Licensed |
| Edition: | Official |
| Language: | en |
| Size: | 1.5 MB |
Included Models
- DocumentAssembler
- Doc2Chunk
- ChunkMapperModel