Description
This pipeline is designed to extract biomarker related entities from text. In this pipeline, 3 NER models and a text matcher are used to extract the biomarkers and their results.
How to use
from sparknlp.pretrained import PretrainedPipeline
ner_pipeline = PretrainedPipeline("explain_clinical_doc_biomarker_light", "en", "clinical/models")
result = ner_pipeline.annotate("""Immunohistochemistry study showed that the tumor cells are positive for epithelial markers-cytokeratin (AE1/AE3) stain, and myoepithelial markers, including cytokeratin 5/6 (CK 5/6), p63, and S100 stains.
Expressions of hormone receptors, including ER, PR, and Her-2/Neu, were all negative.
In the bone- marrow (BM) aspiration, blasts accounted for 88.1% of ANCs, which were positive for CD9 and CD10 on flow cytometry.
Measurements of serum tumor markers showed elevated level of Cyfra21-1: 4.77 ng/mL, NSE: 19.60 ng/mL, and SCCA: 2.58 ng/mL.
Immunohistochemical staining showed positive staining for CK5/6, P40, and negative staining for TTF-1 and weakly positive staining for ALK.
""")
import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline
val ner_pipeline = PretrainedPipeline("explain_clinical_doc_biomarker_light", "en", "clinical/models")
val result = ner_pipeline.annotate("""Immunohistochemistry study showed that the tumor cells are positive for epithelial markers-cytokeratin (AE1/AE3) stain, and myoepithelial markers, including cytokeratin 5/6 (CK 5/6), p63, and S100 stains.
Expressions of hormone receptors, including ER, PR, and Her-2/Neu, were all negative.
In the bone- marrow (BM) aspiration, blasts accounted for 88.1% of ANCs, which were positive for CD9 and CD10 on flow cytometry.
Measurements of serum tumor markers showed elevated level of Cyfra21-1: 4.77 ng/mL, NSE: 19.60 ng/mL, and SCCA: 2.58 ng/mL.
Immunohistochemical staining showed positive staining for CK5/6, P40, and negative staining for TTF-1 and weakly positive staining for ALK.
""")
Results
| | chunks | begin | end | entities |
|---:|:-------------------------------|--------:|------:|:-----------------|
| 0 | positive | 59 | 66 | Biomarker_Result |
| 1 | epithelial markers-cytokeratin | 72 | 101 | Biomarker |
| 2 | AE1/AE3 | 104 | 110 | Biomarker |
| 3 | myoepithelial markers | 124 | 144 | Biomarker |
| 4 | cytokeratin 5/6 | 157 | 171 | Biomarker |
| 5 | CK 5/6 | 174 | 179 | Biomarker |
| 6 | p63 | 183 | 185 | Biomarker |
| 7 | S100 | 192 | 195 | Biomarker |
| 8 | hormone receptors | 221 | 237 | Biomarker |
| 9 | ER | 250 | 251 | Biomarker |
| 10 | PR | 254 | 255 | Biomarker |
| 11 | Her-2/Neu | 262 | 270 | Biomarker |
| 12 | negative | 282 | 289 | Biomarker_Result |
| 13 | positive | 377 | 384 | Biomarker_Result |
| 14 | CD9 | 390 | 392 | Biomarker |
| 15 | CD10 | 398 | 401 | Biomarker |
| 16 | tumor markers | 446 | 458 | Biomarker |
| 17 | elevated level | 467 | 480 | Biomarker_Result |
| 18 | Cyfra21-1 | 485 | 493 | Biomarker |
| 19 | 4.77 ng/mL | 496 | 505 | Biomarker_Result |
| 20 | NSE | 508 | 510 | Biomarker |
| 21 | 19.60 ng/mL | 513 | 523 | Biomarker_Result |
| 22 | SCCA | 530 | 533 | Biomarker |
| 23 | 2.58 ng/mL | 536 | 545 | Biomarker_Result |
| 24 | positive staining | 586 | 602 | Biomarker_Result |
| 25 | CK5/6 | 608 | 612 | Biomarker |
| 26 | P40 | 615 | 617 | Biomarker |
| 27 | negative staining | 624 | 640 | Biomarker_Result |
| 28 | TTF-1 | 646 | 650 | Biomarker |
| 29 | weakly positive staining | 656 | 679 | Biomarker_Result |
| 30 | ALK | 685 | 687 | Biomarker |
Model Information
| Model Name: | explain_clinical_doc_biomarker_light |
| Type: | pipeline |
| Compatibility: | Healthcare NLP 6.0.2+ |
| License: | Licensed |
| Edition: | Official |
| Language: | en |
| Size: | 1.8 GB |
Included Models
- DocumentAssembler
- SentenceDetectorDLModel
- TokenizerModel
- WordEmbeddingsModel
- MedicalNerModel
- NerConverterInternalModel
- MedicalNerModel
- NerConverterInternalModel
- MedicalNerModel
- NerConverterInternalModel
- TextMatcherInternalModel
- ChunkMergeModel