Detect Treatment Entities (TREATMENT)

Description

This pipeline can be used to extract treatments mentioned in medical text.

Predicted Entities

TREATMENT

Copy S3 URI

How to use

from sparknlp.pretrained import PretrainedPipeline

ner_pipeline = PretrainedPipeline("ner_treatment_benchmark_pipeline", "en", "clinical/models")

text = """IN SUMMARY :
The patient was assessed as a 72 year old woman with a background of stage IIIC ovarian carcinoma and documented local recurrence who presents for line 2 of cycle 1 chemotherapy with Adriamycin , Ifex and MESNA .
She also has a low-grade fever of unknown etiology , has a background history of deep venous thrombosis and is therefore currently on anticoagulation and she shows evidence of dehydration and failure to thrive .
It was decided at that time to hold off with the chemotherapy .
The patient was started on Ampicillin and Gentamicin for urinary tract infection which ultimately grew out Escherichia coli sensitive to the above antibiotics and for right lower lobe pneumonia on x-ray .
She was started on nebulizers around-the-clock and chest physical therapy .
On 6/5/94 they were 21 and 2.2 respectively .
Her Coumadin anticoagulation was adjusted to give a prothrombin time between 16 and 18 and an I and R of 2.5-3 .
On June 5 , 1994 it was decided that Mrs. Neathe was not stable enough with a line 2 cycle I chemotherapy with Ifex , Adriamycin and MESNA .
She was therefore well hydrated and was started on her chemotherapy .
In view of her kidney damage it was suggested to change her intravenous antibiotics from Ancef and Gentamicin to Ancef and ciprofloxacin which she tolerated well .
A Neuro-Oncology consult was sought which felt this was probably secondary to Ifex intoxication and her chemotherapy was stopped .
An electroencephalogram was requested and was negative .
No computerized tomography scan or magnetic resonance imaging study of the head was performed .
"""

result = ner_pipeline.fullAnnotate(text)
from sparknlp.pretrained import PretrainedPipeline

ner_pipeline = nlp.PretrainedPipeline("ner_treatment_benchmark_pipeline", "en", "clinical/models")

text = """IN SUMMARY :
The patient was assessed as a 72 year old woman with a background of stage IIIC ovarian carcinoma and documented local recurrence who presents for line 2 of cycle 1 chemotherapy with Adriamycin , Ifex and MESNA .
She also has a low-grade fever of unknown etiology , has a background history of deep venous thrombosis and is therefore currently on anticoagulation and she shows evidence of dehydration and failure to thrive .
It was decided at that time to hold off with the chemotherapy .
The patient was started on Ampicillin and Gentamicin for urinary tract infection which ultimately grew out Escherichia coli sensitive to the above antibiotics and for right lower lobe pneumonia on x-ray .
She was started on nebulizers around-the-clock and chest physical therapy .
On 6/5/94 they were 21 and 2.2 respectively .
Her Coumadin anticoagulation was adjusted to give a prothrombin time between 16 and 18 and an I and R of 2.5-3 .
On June 5 , 1994 it was decided that Mrs. Neathe was not stable enough with a line 2 cycle I chemotherapy with Ifex , Adriamycin and MESNA .
She was therefore well hydrated and was started on her chemotherapy .
In view of her kidney damage it was suggested to change her intravenous antibiotics from Ancef and Gentamicin to Ancef and ciprofloxacin which she tolerated well .
A Neuro-Oncology consult was sought which felt this was probably secondary to Ifex intoxication and her chemotherapy was stopped .
An electroencephalogram was requested and was negative .
No computerized tomography scan or magnetic resonance imaging study of the head was performed .
"""

result = ner_pipeline.fullAnnotate(text)
import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline

val ner_pipeline = PretrainedPipeline("ner_treatment_benchmark_pipeline", "en", "clinical/models")

val text = """IN SUMMARY :
The patient was assessed as a 72 year old woman with a background of stage IIIC ovarian carcinoma and documented local recurrence who presents for line 2 of cycle 1 chemotherapy with Adriamycin , Ifex and MESNA .
She also has a low-grade fever of unknown etiology , has a background history of deep venous thrombosis and is therefore currently on anticoagulation and she shows evidence of dehydration and failure to thrive .
It was decided at that time to hold off with the chemotherapy .
The patient was started on Ampicillin and Gentamicin for urinary tract infection which ultimately grew out Escherichia coli sensitive to the above antibiotics and for right lower lobe pneumonia on x-ray .
She was started on nebulizers around-the-clock and chest physical therapy .
On 6/5/94 they were 21 and 2.2 respectively .
Her Coumadin anticoagulation was adjusted to give a prothrombin time between 16 and 18 and an I and R of 2.5-3 .
On June 5 , 1994 it was decided that Mrs. Neathe was not stable enough with a line 2 cycle I chemotherapy with Ifex , Adriamycin and MESNA .
She was therefore well hydrated and was started on her chemotherapy .
In view of her kidney damage it was suggested to change her intravenous antibiotics from Ancef and Gentamicin to Ancef and ciprofloxacin which she tolerated well .
A Neuro-Oncology consult was sought which felt this was probably secondary to Ifex intoxication and her chemotherapy was stopped .
An electroencephalogram was requested and was negative .
No computerized tomography scan or magnetic resonance imaging study of the head was performed .
"""

val result = ner_pipeline.fullAnnotate(text)

Results

|    | chunk            |   begin |   end | ner_label   |
|---:|:-----------------|--------:|------:|:------------|
|  0 | chemotherapy     |     178 |   189 | TREATMENT   |
|  1 | anticoagulation  |     360 |   374 | TREATMENT   |
|  2 | chemotherapy     |     487 |   498 | TREATMENT   |
|  3 | antibiotics      |     649 |   659 | TREATMENT   |
|  4 | nebulizers       |     726 |   735 | TREATMENT   |
|  5 | physical therapy |     764 |   779 | TREATMENT   |
|  6 | anticoagulation  |     842 |   856 | TREATMENT   |
|  7 | chemotherapy     |    1035 |  1046 | TREATMENT   |
|  8 | chemotherapy     |    1138 |  1149 | TREATMENT   |
|  9 | antibiotics      |    1225 |  1235 | TREATMENT   |
| 10 | chemotherapy     |    1421 |  1432 | TREATMENT   |

Model Information

Model Name: ner_treatment_benchmark_pipeline
Type: pipeline
Compatibility: Healthcare NLP 5.5.3+
License: Licensed
Edition: Official
Language: en
Size: 1.7 GB

Included Models

  • DocumentAssembler
  • SentenceDetector
  • TokenizerModel
  • WordEmbeddingsModel
  • TextMatcherInternalModel
  • MedicalNerModel
  • NerConverterInternalModel
  • MedicalNerModel
  • NerConverterInternalModel
  • MedicalNerModel
  • NerConverterInternalModel
  • ChunkMergeModel
  • ChunkMergeModel

Benchmarking

       label  precision    recall  f1-score   support
           O      0.999     0.999     0.999     82021
   TREATMENT      0.900     0.918     0.909       550
    accuracy      -         -         0.999     82571
   macro-avg      0.950     0.959     0.954     82571
weighted-avg      0.999     0.999     0.999     82571