Description
This model maps vaccine entities to CVX codes using sbiobert_base_cased_mli Sentence Bert Embeddings. Additionally, this model returns status of the vaccine (Active/Inactive/Pending/Non-US) in all_k_aux_labels column.
Predicted Entities
CVX Code, Status
How to use
documentAssembler = DocumentAssembler()\
.setInputCol("text")\
.setOutputCol("ner_chunk")
sbert_embedder = BertSentenceEmbeddings.pretrained("sbiobert_base_cased_mli", "en", "clinical/models")\
.setInputCols(["ner_chunk"])\
.setOutputCol("sbert_embeddings")
cvx_resolver = SentenceEntityResolverModel.pretrained("sbiobertresolve_cvx", "en", "clinical/models")\
.setInputCols(["sbert_embeddings"])\
.setOutputCol("cvx_code")\
.setDistanceFunction("EUCLIDEAN")
pipelineModel = PipelineModel( stages = [ documentAssembler, sbert_embedder, cvx_resolver ])
light_model = LightPipeline(pipelineModel)
result = light_model.fullAnnotate(["Sinovac", "Moderna", "BIOTHRAX"])
val documentAssembler = new DocumentAssembler()
.setInputCol("text")
.setOutputCol("ner_chunk")
val sbert_embedder = BertSentenceEmbeddings.pretrained("sbiobert_base_cased_mli", "en","clinical/models")
.setInputCols(Array("ner_chunk"))
.setOutputCol("sbert_embeddings")
val cvx_resolver = SentenceEntityResolverModel.pretrained("sbiobertresolve_cvx", "en", "clinical/models")
.setInputCols(Array("sbert_embeddings"))
.setOutputCol("cvx_code")
.setDistanceFunction("EUCLIDEAN")
val cvx_pipelineModel = new PipelineModel().setStages(Array(documentAssembler, sbert_embedder, cvx_resolver))
val light_model = LightPipeline(cvx_pipelineModel)
val result = light_model.fullAnnotate(Array("Sinovac", "Moderna", "BIOTHRAX"))
import nlu
nlu.load("en.resolve.cvx").predict("""Put your text here.""")
Results
+----------+--------+-------------------------------------------------------+--------+
|ner_chunk |cvx_code|resolved_text |Status |
+----------+--------+-------------------------------------------------------+--------+
|Sinovac |511 |COVID-19 IV Non-US Vaccine (CoronaVac, Sinovac) |Non-US |
|Moderna |227 |COVID-19, mRNA, LNP-S, PF, pediatric 50 mcg/0.5 mL dose|Inactive|
|BIOTHRAX |24 |anthrax |Active |
+----------+--------+-------------------------------------------------------+--------+
Model Information
| Model Name: | sbiobertresolve_cvx |
| Compatibility: | Healthcare NLP 4.2.1+ |
| License: | Licensed |
| Edition: | Official |
| Input Labels: | [bert_embeddings] |
| Output Labels: | [cvx_code] |
| Language: | en |
| Size: | 1.6 MB |
| Case sensitive: | false |