Description
This pipeline includes Named-Entity Recognition, Assertion Status and Relation Extraction models to extract information from oncology texts. This pipeline extracts diagnoses, treatments, tests, anatomical references and demographic entities.
Predicted Entities
How to use
from sparknlp.pretrained import PretrainedPipeline
pipeline = PretrainedPipeline("oncology_general_pipeline", "en", "clinical/models")
text = '''The patient underwent a left mastectomy for a left breast cancer two months ago.
The tumor is positive for ER and PR.'''
result = pipeline.fullAnnotate(text)
import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline
val pipeline = new PretrainedPipeline("oncology_general_pipeline", "en", "clinical/models")
val text = "The patient underwent a left mastectomy for a left breast cancer two months ago.
The tumor is positive for ER and PR."
val result = pipeline.fullAnnotate(text)
import nlu
nlu.load("en.oncology_general.pipeline").predict("""The patient underwent a left mastectomy for a left breast cancer two months ago.
The tumor is positive for ER and PR.""")
Results
******************** ner_oncology_wip results ********************
| chunk | ner_label |
|:---------------|:-----------------|
| left | Direction |
| mastectomy | Cancer_Surgery |
| left | Direction |
| breast cancer | Cancer_Dx |
| two months ago | Relative_Date |
| tumor | Tumor_Finding |
| positive | Biomarker_Result |
| ER | Biomarker |
| PR | Biomarker |
******************** ner_oncology_diagnosis_wip results ********************
| chunk | ner_label |
|:--------------|:--------------|
| breast cancer | Cancer_Dx |
| tumor | Tumor_Finding |
******************** ner_oncology_tnm_wip results ********************
| chunk | ner_label |
|:--------------|:------------|
| breast cancer | Cancer_Dx |
| tumor | Tumor |
******************** ner_oncology_therapy_wip results ********************
| chunk | ner_label |
|:-----------|:---------------|
| mastectomy | Cancer_Surgery |
******************** ner_oncology_test_wip results ********************
| chunk | ner_label |
|:---------|:-----------------|
| positive | Biomarker_Result |
| ER | Biomarker |
| PR | Biomarker |
******************** assertion_oncology_wip results ********************
| chunk | ner_label | assertion |
|:--------------|:---------------|:------------|
| mastectomy | Cancer_Surgery | Past |
| breast cancer | Cancer_Dx | Present |
| tumor | Tumor_Finding | Present |
| ER | Biomarker | Present |
| PR | Biomarker | Present |
******************** re_oncology_wip results ********************
| chunk1 | entity1 | chunk2 | entity2 | relation |
|:--------------|:-----------------|:---------------|:--------------|:--------------|
| mastectomy | Cancer_Surgery | two months ago | Relative_Date | is_related_to |
| breast cancer | Cancer_Dx | two months ago | Relative_Date | is_related_to |
| tumor | Tumor_Finding | ER | Biomarker | O |
| tumor | Tumor_Finding | PR | Biomarker | O |
| positive | Biomarker_Result | ER | Biomarker | is_related_to |
| positive | Biomarker_Result | PR | Biomarker | is_related_to |
******************** re_oncology_granular_wip results ********************
| chunk1 | entity1 | chunk2 | entity2 | relation |
|:--------------|:-----------------|:---------------|:--------------|:--------------|
| mastectomy | Cancer_Surgery | two months ago | Relative_Date | is_date_of |
| breast cancer | Cancer_Dx | two months ago | Relative_Date | is_date_of |
| tumor | Tumor_Finding | ER | Biomarker | O |
| tumor | Tumor_Finding | PR | Biomarker | O |
| positive | Biomarker_Result | ER | Biomarker | is_finding_of |
| positive | Biomarker_Result | PR | Biomarker | is_finding_of |
Model Information
| Model Name: | oncology_general_pipeline |
| Type: | pipeline |
| Compatibility: | Healthcare NLP 4.4.4+ |
| License: | Licensed |
| Edition: | Official |
| Language: | en |
| Size: | 1.7 GB |
Included Models
- DocumentAssembler
- SentenceDetectorDLModel
- TokenizerModel
- WordEmbeddingsModel
- MedicalNerModel
- NerConverter
- MedicalNerModel
- NerConverter
- MedicalNerModel
- NerConverter
- MedicalNerModel
- NerConverter
- MedicalNerModel
- NerConverter
- ChunkMergeModel
- ChunkMergeModel
- AssertionDLModel
- PerceptronModel
- DependencyParserModel
- RelationExtractionModel
- RelationExtractionModel