Description
A pipeline for detecting posology entities with the ner_posology_large NER model, assigning their assertion status with assertion_oncology_wip model, and extracting relations between posology-related terminology with posology_re relation extraction model.
Predicted Entities
DRUG, DOSAGE, DURATION, FORM, FREQUENCY, ROUTE, STRENGTH
How to use
from sparknlp.pretrained import PretrainedPipeline
ner_pipeline = PretrainedPipeline("explain_clinical_doc_medication", "en", "clinical/models")
result = ner_pipeline.annotate("""The patient is a 30-year-old female with diabetes mellitus type 2. She received a course of Bactrim for 14 days for UTI.
She was prescribed 5000 units of Fragmin subcutaneously daily. She was also prescribed 40 units of Lantus subcutaneously at night.""")
Results
# ner_chunk
+--------------+-----+---+---------+
| ner_chunk|begin|end|ner_label|
+--------------+-----+---+---------+
| Bactrim| 92| 98| DRUG|
| for 14 days| 100|110| DURATION|
| 5000 units| 141|150| DOSAGE|
| Fragmin| 155|161| DRUG|
|subcutaneously| 163|176| ROUTE|
| daily| 178|182|FREQUENCY|
| 40 units| 209|216| DOSAGE|
| Lantus| 221|226| DRUG|
|subcutaneously| 228|241| ROUTE|
| at night| 243|250|FREQUENCY|
+--------------+-----+---+---------+
# assertion
+-------+-----+---+--------+---------+-----------+
| chunks|begin|end|entities|assertion|confidence)|
+-------+-----+---+--------+---------+-----------+
|Bactrim| 92| 98| DRUG| Past| 0.9324|
|Fragmin| 155|161| DRUG| Present| 0.7456|
| Lantus| 190|195| DRUG| Present| 0.4984|
+-------+-----+---+--------+---------+-----------+
# relation
+--------+--------------+---------+----------+-------+--------------+---------+----------+
|sentence| relation|direction| chunk1|entity1| chunk2| entity2|confidence|
+--------+--------------+---------+----------+-------+--------------+---------+----------+
| 1| DRUG-DURATION| both| Bactrim| DRUG| for 14 days| DURATION| 1.0|
| 2| DOSAGE-DRUG| both|5000 units| DOSAGE| Fragmin| DRUG| 1.0|
| 2| DRUG-ROUTE| both| Fragmin| DRUG|subcutaneously| ROUTE| 1.0|
| 2|DRUG-FREQUENCY| both| Fragmin| DRUG| daily|FREQUENCY| 1.0|
| 3| DOSAGE-DRUG| both| 40 units| DOSAGE| Lantus| DRUG| 1.0|
| 3| DRUG-ROUTE| both| Lantus| DRUG|subcutaneously| ROUTE| 1.0|
| 3|DRUG-FREQUENCY| both| Lantus| DRUG| at night|FREQUENCY| 1.0|
+--------+--------------+---------+----------+-------+--------------+---------+----------+
Model Information
| Model Name: | explain_clinical_doc_medication |
| Type: | pipeline |
| Compatibility: | Healthcare NLP 5.3.3+ |
| License: | Licensed |
| Edition: | Official |
| Language: | en |
| Size: | 1.7 GB |
Included Models
- DocumentAssembler
- SentenceDetectorDLModel
- TokenizerModel
- WordEmbeddingsModel
- MedicalNerModel
- NerConverter
- NerConverterInternalModel
- AssertionDLModel
- PerceptronModel
- DependencyParserModel
- PosologyREModel