Description
This model maps clinical entities and concepts to 4 major categories of UMLS CUI codes using sbiobert_base_cased_mli Sentence Bert Embeddings. It has faster load time, with a speedup of about 6X when compared to previous versions. Also the load process now is more memory friendly meaning that the maximum memory required during load time is smaller, reducing the chances of OOM exceptions, and thus relaxing hardware requirements.
Predicted Entities
This model returns CUI (concept unique identifier) codes for Clinical Findings, Medical Devices, Anatomical Structures and Injuries & Poisoning terms
Live Demo Open in Colab Copy S3 URI
How to use
sbiobertresolve_umls_major_concepts resolver model must be used with sbiobert_base_cased_mli as embeddings ner_jsl as NER model. Cerebrovascular_Disease, Communicable_Disease, Diabetes, Disease_Syndrome_Disorder, Heart_Disease, Hyperlipidemia, Hypertension, Injury_or_Poisoning, Kidney_Disease, Medical-Device, Obesity, Oncological, Overweight, Psychological_Condition, Symptom, VS_Finding, ImagingFindings, EKG_Findings set in .setWhiteList().
...
chunk2doc = Chunk2Doc().setInputCols("ner_chunk").setOutputCol("ner_chunk_doc")
sbert_embedder = BertSentenceEmbeddings\
.pretrained("sbiobert_base_cased_mli","en","clinical/models")\
.setInputCols(["ner_chunk_doc"])\
.setOutputCol("sbert_embeddings")
resolver = SentenceEntityResolverModel\
.pretrained("sbiobertresolve_umls_major_concepts", "en", "clinical/models") \
.setInputCols(["ner_chunk", "sbert_embeddings"]) \
.setOutputCol("resolution")\
.setDistanceFunction("EUCLIDEAN")
pipeline = Pipeline(stages = [document_assembler, sentence_detector, tokens, embeddings, ner, ner_converter, chunk2doc, sbert_embedder, resolver])
data = spark.createDataFrame([["""A 28-year-old female with a history of gestational diabetes mellitus diagnosed eight years prior to presentation and subsequent type two diabetes mellitus (T2DM), one prior episode of HTG-induced pancreatitis three years prior to presentation, associated with an acute hepatitis, and obesity with a body mass index (BMI) of 33.5 kg/m2, presented with a one-week history of polyuria, polydipsia, poor appetite, and vomiting."""]]).toDF("text")
results = pipeline.fit(data).transform(data)
import nlu
nlu.load("en.resolve.umls").predict("""A 28-year-old female with a history of gestational diabetes mellitus diagnosed eight years prior to presentation and subsequent type two diabetes mellitus (T2DM), one prior episode of HTG-induced pancreatitis three years prior to presentation, associated with an acute hepatitis, and obesity with a body mass index (BMI) of 33.5 kg/m2, presented with a one-week history of polyuria, polydipsia, poor appetite, and vomiting.""")
Results
| | ner_chunk | resolution |
|---:|:------------------------------|:-------------|
| 0 | 28-year-old | C1864118 |
| 1 | female | C3887375 |
| 2 | gestational diabetes mellitus | C2183115 |
| 3 | eight years prior | C5195266 |
| 4 | subsequent | C3844350 |
| 5 | type two diabetes mellitus | C4014362 |
| 6 | T2DM | C4014362 |
| 7 | HTG-induced pancreatitis | C4554179 |
| 8 | three years prior | C1866782 |
| 9 | acute | C1332147 |
| 10 | hepatitis | C1963279 |
| 11 | obesity | C1963185 |
| 12 | body mass index | C0578022 |
| 13 | 33.5 kg/m2 | C2911054 |
| 14 | one-week | C0420331 |
| 15 | polyuria | C3278312 |
| 16 | polydipsia | C3278316 |
| 17 | poor appetite | C0541799 |
| 18 | vomiting | C1963281 |
Model Information
| Model Name: | sbiobertresolve_umls_major_concepts |
| Compatibility: | Healthcare NLP 3.0.4+ |
| License: | Licensed |
| Edition: | Official |
| Input Labels: | [sentence_embeddings] |
| Output Labels: | [umls_code] |
| Language: | en |
| Case sensitive: | false |