RE Pipeline between Problem, Test, and Findings in Reports

Description

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

Predicted Entities

Admission_Discharge, Age, Alcohol, Allergen, BMI, Birth_Entity, Blood_Pressure, Cerebrovascular_Disease, Clinical_Dept, Communicable_Disease, Date, Death_Entity, Diabetes, Diet, Direction, Disease_Syndrome_Disorder, Dosage, Drug, Duration, EKG_Findings, Employment, External_body_part_or_region, Family_History_Header, Fetus_NewBorn, Form, Frequency, Gender, HDL, Heart_Disease, Height, Hyperlipidemia, Hypertension, ImagingFindings, Imaging_Technique, Injury_or_Poisoning, Internal_organ_or_component, Kidney_Disease, LDL, Labour_Delivery, Medical_Device, Medical_History_Header, Modifier, O2_Saturation, Obesity, Oncological, Overweight, Oxygen_Therapy, Pregnancy, Procedure, Psychological_Condition, Pulse, Race_Ethnicity, Relationship_Status, RelativeDate, RelativeTime, Respiration, Route, Section_Header, Sexually_Active_or_Sexual_Orientation, Smoking, Social_History_Header, Strength, Substance, Substance_Quantity, Symptom, Temperature, Test, Test_Result, Time, Total_Cholesterol, Treatment, Triglycerides, VS_Finding, Vaccine, Vital_Signs_Header, Weight

Live Demo Open in Colab Download Copy S3 URI

How to use

from sparknlp.pretrained import PretrainedPipeline

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

pipeline.fullAnnotate("Targeted biopsy of this lesion for histological correlation should be considered.")

import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline

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

pipeline.fullAnnotate("Targeted biopsy of this lesion for histological correlation should be considered.")

import nlu
nlu.load("en.relation.test_problem_finding.pipeline").predict("""Targeted biopsy of this lesion for histological correlation should be considered.""")

Results

| index | relations    | entity1      | chunk1              | entity2      |  chunk2 |
|-------|--------------|--------------|---------------------|--------------|---------|
| 0     | 1            | PROCEDURE    | biopsy              | SYMPTOM      |  lesion | 

Model Information

Included Models

• DocumentAssembler
• SentenceDetector
• TokenizerModel
• PerceptronModel
• WordEmbeddingsModel
• MedicalNerModel
• NerConverter
• DependencyParserModel
• RelationExtractionModel