Mapping MedDRA-LLT (Lowest Level Term) Codes With Their Corresponding MedDRA-PT (Preferred Term) Codes

Description

This pretrained model maps MedDRA LLT (Lowest Level Term) codes to their corresponding MedDRA PT (Preferred Term) codes.

Predicted Entities

How to use

document_assembler = DocumentAssembler()\
      .setInputCol('text')\
      .setOutputCol('doc')

chunk_assembler = Doc2Chunk()\
      .setInputCols(['doc'])\
      .setOutputCol('chunk')
 
mapperModel = ChunkMapperModel.load('meddra_llt_pt_mapper')\
    .setInputCols(["chunk"])\
    .setOutputCol("mappings")\
    .setRels(["pt_code"])


pipeline = Pipeline(stages=[
    document_assembler,
    chunk_assembler,
    mapperModel
])

data = spark.createDataFrame([["10002442"], ["10000007"], ["10003696"]]).toDF("text")

mapper_model = pipeline.fit(data)
result = mapper_model.transform(data)
val document_assembler = DocumentAssembler()
      .setInputCol("text")
      .setOutputCol("doc")

val chunk_assembler = Doc2Chunk()
      .setInputCols(Array("doc"))
      .setOutputCol("chunk")
 
val mapperModel = ChunkMapperModel.load("meddra_llt_pt_mapper")
    .setInputCols(Array("chunk"))
    .setOutputCol("mappings")
    .setRels(["pt_code"])


val pipeline = new Pipeline().setStages(Array(
    document_assembler,
    chunk_assembler,
    mapperModel))

val data = Seq("10002442", "10000007", "10003696").toDF("text")

val mapper_model = pipeline.fit(data)
val result = mapper_model.transform(data)

Results

+--------+----------------------------------------+
|llt_code|pt_code                                 |
+--------+----------------------------------------+
|10002442|10002442:Angiogram pulmonary normal     |
|10000007|10000007:17 ketosteroids urine decreased|
|10003696|10001324:Adrenal atrophy                |
+--------+----------------------------------------+

Model Information

Model Name: meddra_llt_pt_mapper
Compatibility: Healthcare NLP 5.4.1+
License: Licensed
Edition: Official
Input Labels: [ner_chunk]
Output Labels: [mappings]
Language: en
Size: 1.9 MB

References

This model is trained with the September 2024 (v27.1) release of MedDRA dataset.

To utilize this model, possession of a valid MedDRA license is requisite. If you possess one and wish to use this model, kindly contact us at support@johnsnowlabs.com.