Description
This pipeline detects drugs, dosage, form, frequency, duration, route, and drug strength in text.
Predicted Entities
DRUG, STRENGTH, DURATION, FREQUENCY, FORM, DOSAGE, ROUTE.
Live Demo Open in Colab Copy S3 URI
How to use
from sparknlp.pretrained import PretrainedPipeline
pipeline = PretrainedPipeline('recognize_entities_posology', 'en', 'clinical/models')
annotations = pipeline.fullAnnotate("""The patient was perscriped 50MG penicilin for is headache""")[0]
annotations.keys()
val pipeline = new PretrainedPipeline("recognize_entities_posology", "en", "clinical/models")
val result = pipeline.fullAnnotate("""The patient was perscriped 50MG penicilin for is headache""")(0)
import nlu
result_df = nlu.load('ner.posology').predict("""The patient was perscriped 50MG penicilin for is headache""")
result_df
Results
+-----------------------------------------+
|result |
+-----------------------------------------+
|[O, O, O, O, B-Strength, B-Drug, O, O, O]|
+-----------------------------------------+
+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
|ner |
+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
|[[named_entity, 0, 2, O, [word -> The, confidence -> 1.0], []], [named_entity, 4, 10, O, [word -> patient, confidence -> 0.9993], []], [named_entity, 12, 14, O, [word -> was, confidence -> 1.0], []], [named_entity, 16, 25, O, [word -> perscriped, confidence -> 0.9985], []], [named_entity, 27, 30, B-Strength, [word -> 50MG, confidence -> 0.9966], []], [named_entity, 32, 40, B-Drug, [word -> penicilin, confidence -> 0.9934], []], [named_entity, 42, 44, O, [word -> for, confidence -> 0.9999], []], [named_entity, 46, 47, O, [word -> is, confidence -> 0.9468], []], [named_entity, 49, 56, O, [word -> headache, confidence -> 0.9805], []]]|
+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
Model Information
| Model Name: | recognize_entities_posology |
| Type: | pipeline |
| Compatibility: | Spark NLP for Healthcare 3.0.0+ |
| License: | Licensed |
| Edition: | Official |
| Language: | en |
Included Models
- DocumentAssembler
- SentenceDetector
- TokenizerModel
- WordEmbeddingsModel
- NerDLModel
- NerConverter