Edit on GitHub
  • John Snow Labs
  • Sep 11, 2026

Sentence Entity Resolver for SNOMED CT (Clinical Findings) (bge_base_en_v1_5_onnx embeddings)

  • en
  • snomed
  • resolver
  • licensed
  • clinical
  • findings
  • bge

Description

This model maps extracted clinical NER entities to SNOMED CT concepts using bge_base_en_v1_5_onnx embeddings.

It is trained on SNOMED CT US Edition 20260901 release.

Live Demo Open in Colab Copy S3 URI

How to use

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

sentenceDetectorDL = SentenceDetectorDLModel.pretrained("sentence_detector_dl_healthcare", "en", "clinical/models")\
    .setInputCols(["document"])\
    .setOutputCol("sentence")

tokenizer = Tokenizer()\
    .setInputCols(["sentence"])\
    .setOutputCol("token")

word_embeddings = WordEmbeddingsModel.pretrained("embeddings_clinical","en","clinical/models")\
    .setInputCols(["sentence","token"])\
    .setOutputCol("embeddings")

ner_model = MedicalNerModel.pretrained("ner_jsl","en","clinical/models")\
    .setInputCols(["sentence","token","embeddings"])\
    .setOutputCol("ner_tags")

ner_converter = NerConverterInternal()\
    .setInputCols(["sentence","token","ner_tags"])\
    .setOutputCol("ner_chunk")\
    .setWhiteList(["Kidney_Disease", "Cerebrovascular_Disease", "Heart_Disease", "Disease_Syndrome_Disorder", "ImagingFindings", "Symptom", "VS_Finding", "EKG_Findings", "Communicable_Disease"])

chunk2doc = Chunk2Doc()\
    .setInputCols(["ner_chunk"])\
    .setOutputCol("ner_chunk_doc")

embedder = BGEEmbeddings.pretrained("bge_base_en_v1_5_onnx", "en")\
    .setInputCols(["ner_chunk_doc"])\
    .setOutputCol("bge_embeddings")\
    .setCaseSensitive(False)

resolver = SentenceEntityResolverModel.pretrained("bgeresolve_snomed_findings_20260901","en","clinical/models")\
    .setInputCols(["bge_embeddings"])\
    .setOutputCol("snomed_code")\
    .setDistanceFunction("EUCLIDEAN")\
    .setThreshold(1000)

pipeline = Pipeline(stages=[\
    documentAssembler, sentenceDetectorDL, tokenizer, word_embeddings, ner_model, ner_converter, chunk2doc, embedder, resolver\
])

data = spark.createDataFrame([["The patient presented with recurrent fevers. Clinically she appeared cachectic with hepatosplenomegaly. Laboratory results confirmed pancytopenia."]]).toDF("text")
result = pipeline.fit(data).transform(data)

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

sentenceDetectorDL = nlp.SentenceDetectorDLModel.pretrained("sentence_detector_dl_healthcare", "en", "clinical/models")\
    .setInputCols(["document"])\
    .setOutputCol("sentence")

tokenizer = nlp.Tokenizer()\
    .setInputCols(["sentence"])\
    .setOutputCol("token")

word_embeddings = nlp.WordEmbeddingsModel.pretrained("embeddings_clinical","en","clinical/models")\
    .setInputCols(["sentence","token"])\
    .setOutputCol("embeddings")

ner_model = medical.NerModel.pretrained("ner_jsl","en","clinical/models")\
    .setInputCols(["sentence","token","embeddings"])\
    .setOutputCol("ner_tags")

ner_converter = medical.NerConverterInternal()\
    .setInputCols(["sentence","token","ner_tags"])\
    .setOutputCol("ner_chunk")\
    .setWhiteList(["Kidney_Disease", "Cerebrovascular_Disease", "Heart_Disease", "Disease_Syndrome_Disorder", "ImagingFindings", "Symptom", "VS_Finding", "EKG_Findings", "Communicable_Disease"])

chunk2doc = nlp.Chunk2Doc()\
    .setInputCols(["ner_chunk"])\
    .setOutputCol("ner_chunk_doc")

embedder = nlp.BGEEmbeddings.pretrained("bge_base_en_v1_5_onnx", "en")\
    .setInputCols(["ner_chunk_doc"])\
    .setOutputCol("bge_embeddings")\
    .setCaseSensitive(False)

resolver = medical.SentenceEntityResolverModel.pretrained("bgeresolve_snomed_findings_20260901","en","clinical/models")\
    .setInputCols(["bge_embeddings"])\
    .setOutputCol("snomed_code")\
    .setDistanceFunction("EUCLIDEAN")\
    .setThreshold(1000)

pipeline = nlp.Pipeline(stages=[\
    documentAssembler, sentenceDetectorDL, tokenizer, word_embeddings, ner_model, ner_converter, chunk2doc, embedder, resolver\
])

data = spark.createDataFrame([["The patient presented with recurrent fevers. Clinically she appeared cachectic with hepatosplenomegaly. Laboratory results confirmed pancytopenia."]]).toDF("text")
result = pipeline.fit(data).transform(data)


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

val sentenceDetectorDL = SentenceDetectorDLModel
    .pretrained("sentence_detector_dl_healthcare", "en", "clinical/models")
    .setInputCols(Array("document"))
    .setOutputCol("sentence")

val tokenizer = new Tokenizer()
    .setInputCols("sentence")
    .setOutputCol("token")

val word_embeddings = WordEmbeddingsModel
    .pretrained("embeddings_clinical", "en", "clinical/models")
    .setInputCols(Array("sentence", "token"))
    .setOutputCol("embeddings")

val ner_model = MedicalNerModel
    .pretrained("ner_jsl", "en", "clinical/models")
    .setInputCols(Array("sentence", "token", "embeddings"))
    .setOutputCol("ner_tags")

val ner_converter = new NerConverterInternal()
    .setInputCols(Array("sentence", "token", "ner_tags"))
    .setOutputCol("ner_chunk")
    .setWhiteList(Array("Kidney_Disease", "Cerebrovascular_Disease", "Heart_Disease", "Disease_Syndrome_Disorder", "ImagingFindings", "Symptom", "VS_Finding", "EKG_Findings", "Communicable_Disease"))

val chunk2doc = new Chunk2Doc()
    .setInputCols(Array("ner_chunk"))
    .setOutputCol("ner_chunk_doc")

val embedder = BGEEmbeddings
    .pretrained("bge_base_en_v1_5_onnx", "en")
    .setInputCols(Array("ner_chunk_doc"))
    .setOutputCol("bge_embeddings")
    .setCaseSensitive(false)

val resolver = SentenceEntityResolverModel
    .pretrained("bgeresolve_snomed_findings_20260901", "en", "clinical/models")
    .setInputCols(Array("bge_embeddings"))
    .setOutputCol("snomed_code")
    .setDistanceFunction("EUCLIDEAN")
    .setThreshold(1000)

val pipeline = new Pipeline().setStages(Array(
    documentAssembler, sentenceDetectorDL, tokenizer, word_embeddings, ner_model, ner_converter, chunk2doc, embedder, resolver
))

val data = Seq("The patient presented with recurrent fevers. Clinically she appeared cachectic with hepatosplenomegaly. Laboratory results confirmed pancytopenia.").toDF("text")
val res = pipeline.fit(data).transform(data)

Results

| chunk              | label      |   snomed_code | resolution         | all_codes                                                                           | all_resolutions                                                                     |
|:-------------------|:-----------|--------------:|:-------------------|:------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
| fevers             | VS_Finding |     386661006 | fever              | 386661006:::416113008:::722892007:::274640006:::186694006:::135882008:::10151000... | fever:::febrile illness (fever):::fever with infection:::chills and fever:::swea... |
| cachectic          | Symptom    |     238108007 | cachectic          | 238108007:::201139004:::788876001:::110988004:::201275005:::284529003:::24012800... | cachectic:::cachectic alopecia:::malignant cachexia:::melanoderma cachecticorum:... |
| hepatosplenomegaly | Symptom    |      36760000 | hepatosplenomegaly | 36760000:::80378000:::275598004:::403779009:::16294009:::80515008:::714254003:::... | hepatosplenomegaly:::neonatal hepatosplenomegaly:::hepatosplenomegalic lipoidosi... |
| pancytopenia       | Symptom    |     127034005 | pancytopenia       | 127034005:::183005:::234367000:::50820005:::768556005:::736024007:::38970002:::2... | pancytopenia:::autoimmune pancytopenia:::pancytopenia with pancreatitis:::cytope... |

Model Information

Model Name: bgeresolve_snomed_findings_20260901
Compatibility: Healthcare NLP 6.4.1+
License: Licensed
Edition: Official
Input Labels: [bge_embeddings]
Output Labels: [snomed_code]
Language: en
Size: 688.7 MB
Case sensitive: false