NER Pipeline for Anatomy Entities - Voice of the Patient

Description

This pipeline extracts mentions of anatomical sites from health-related text in colloquial language.

Predicted Entities

BodyPart, Laterality

Live Demo Open in Colab Copy S3 URI

How to use

from sparknlp.pretrained import PretrainedPipeline

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

pipeline.annotate("
Ugh, I pulled a muscle in my neck from sleeping weird last night. It's like a knot in my trapezius and it hurts to turn my head.
")
import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline

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

val result = pipeline.annotate("
Ugh, I pulled a muscle in my neck from sleeping weird last night. It's like a knot in my trapezius and it hurts to turn my head.
")

Results

| chunk     | ner_label   |
|:----------|:------------|
| muscle    | BodyPart    |
| neck      | BodyPart    |
| trapezius | BodyPart    |
| head      | BodyPart    |

Model Information

Model Name: ner_vop_anatomy_pipeline
Type: pipeline
Compatibility: Healthcare NLP 4.4.3+
License: Licensed
Edition: Official
Language: en
Size: 791.6 MB

Included Models

  • DocumentAssembler
  • SentenceDetectorDLModel
  • TokenizerModel
  • WordEmbeddingsModel
  • MedicalNerModel
  • NerConverter