Pipeline to Detect Clinical Events (Admissions)

Description

This pretrained pipeline is built on the top of ner_events_admission_clinical model.

Predicted Entities

ADMISSION, CLINICAL_DEPT, DATE, DISCHARGE, DURATION, EVIDENTIAL, FREQUENCY, OCCURRENCE, PROBLEM, TEST, TIME, TREATMENT

Live Demo Open in Colab Download Copy S3 URI

How to use

from sparknlp.pretrained import PretrainedPipeline

pipeline = PretrainedPipeline("ner_events_admission_clinical_pipeline", "en", "clinical/models")

text = '''The patient presented to the emergency room last evening.'''

result = pipeline.fullAnnotate(text)

import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline

val pipeline = new PretrainedPipeline("ner_events_admission_clinical_pipeline", "en", "clinical/models")

val text = "The patient presented to the emergency room last evening."

val result = pipeline.fullAnnotate(text)

import nlu
nlu.load("en.med_ner.events_admission_clinical.pipeline").predict("""The patient presented to the emergency room last evening.""")

Results

|    | ner_chunk          |   begin |   end | ner_label     |   confidence |
|---:|:-------------------|--------:|------:|:--------------|-------------:|
|  0 | presented          |      12 |    20 | OCCURRENCE    |       0.6219 |
|  1 | the emergency room |      25 |    42 | CLINICAL_DEPT |       0.812  |
|  2 | last evening       |      44 |    55 | TIME          |       0.9534 |

Model Information

Included Models

• DocumentAssembler
• SentenceDetectorDLModel
• TokenizerModel
• WordEmbeddingsModel
• MedicalNerModel
• NerConverterInternalModel