Description
A pipeline for detecting radiology entities with the ner_radiology NER model, assigning their assertion status with assertion_dl_radiology model, and extracting relations between the diagnosis, test, and findings with re_test_problem_finding relation extraction model.
How to use
from sparknlp.pretrained import PretrainedPipeline
pipeline = PretrainedPipeline("explain_clinical_doc_radiology", "en", "clinical/models")
result = pipeline.fullAnnotate("""Bilateral breast ultrasound was subsequently performed, which demonstrated an ovoid mass measuring approximately 0.5 x 0.5 x 0.4 cm in diameter located within the anteromedial aspect of the left shoulder. This mass demonstrates isoechoic echotexture to the adjacent muscle, with no evidence of internal color flow. This may represent benign fibrous tissue or a lipoma.""")[0]
import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline
val pipeline = new PretrainedPipeline("explain_clinical_doc_radiology", "en", "clinical/models")
val result = pipeline.fullAnnotate("""Bilateral breast ultrasound was subsequently performed, which demonstrated an ovoid mass measuring approximately 0.5 x 0.5 x 0.4 cm in diameter located within the anteromedial aspect of the left shoulder. This mass demonstrates isoechoic echotexture to the adjacent muscle, with no evidence of internal color flow. This may represent benign fibrous tissue or a lipoma.""")(0)
import nlu
nlu.load("en.explain_doc.clinical_radiology.pipeline").predict("""Bilateral breast ultrasound was subsequently performed, which demonstrated an ovoid mass measuring approximately 0.5 x 0.5 x 0.4 cm in diameter located within the anteromedial aspect of the left shoulder. This mass demonstrates isoechoic echotexture to the adjacent muscle, with no evidence of internal color flow. This may represent benign fibrous tissue or a lipoma.""")
Results
+----+------------------------------------------+---------------------------+
| | chunks | entities |
|---:|:-----------------------------------------|:--------------------------|
| 0 | Bilateral breast | BodyPart |
| 1 | ultrasound | ImagingTest |
| 2 | ovoid mass | ImagingFindings |
| 3 | 0.5 x 0.5 x 0.4 | Measurements |
| 4 | cm | Units |
| 5 | anteromedial aspect of the left shoulder | BodyPart |
| 6 | mass | ImagingFindings |
| 7 | isoechoic echotexture | ImagingFindings |
| 8 | muscle | BodyPart |
| 9 | internal color flow | ImagingFindings |
| 10 | benign fibrous tissue | ImagingFindings |
| 11 | lipoma | Disease_Syndrome_Disorder |
+----+------------------------------------------+---------------------------+
+----+-----------------------+---------------------------+-------------+
| | chunks | entities | assertion |
|---:|:----------------------|:--------------------------|:------------|
| 0 | ultrasound | ImagingTest | Confirmed |
| 1 | ovoid mass | ImagingFindings | Confirmed |
| 2 | mass | ImagingFindings | Confirmed |
| 3 | isoechoic echotexture | ImagingFindings | Confirmed |
| 4 | internal color flow | ImagingFindings | Negative |
| 5 | benign fibrous tissue | ImagingFindings | Suspected |
| 6 | lipoma | Disease_Syndrome_Disorder | Suspected |
+----+-----------------------+---------------------------+-------------+
+---------+-----------------+-----------------------+---------------------------+------------+
|relation | entity1 | chunk1 | entity2 | chunk2 |
|--------:|:----------------|:----------------------|:--------------------------|:-----------|
| 1 | ImagingTest | ultrasound | ImagingFindings | ovoid mass |
| 0 | ImagingFindings | benign fibrous tissue | Disease_Syndrome_Disorder | lipoma |
+---------+-----------------+-----------------------+---------------------------+------------+
Model Information
| Model Name: | explain_clinical_doc_radiology |
| Type: | pipeline |
| Compatibility: | Healthcare NLP 4.0.0+ |
| License: | Licensed |
| Edition: | Official |
| Language: | en |
| Size: | 1.7 GB |
Included Models
- DocumentAssembler
- SentenceDetectorDLModel
- TokenizerModel
- WordEmbeddingsModel
- MedicalNerModel
- NerConverter
- NerConverter
- AssertionDLModel
- PerceptronModel
- DependencyParserModel
- RelationExtractionModel