Description
This model detects the assertion status of entities related to cancer diagnosis (including Metastasis, Cancer_Dx and Tumor_Finding, among others).
Predicted Entities
Absent, Family, Hypothetical, Possible, Present
Live Demo Open in Colab Copy S3 URI
How to use
document_assembler = DocumentAssembler()\
.setInputCol("text")\
.setOutputCol("document")
sentence_detector = SentenceDetectorDLModel.pretrained("sentence_detector_dl_healthcare","en","clinical/models")\
.setInputCols(["document"])\
.setOutputCol("sentence")
tokenizer = Tokenizer() \
.setInputCols(["sentence"]) \
.setOutputCol("token")
word_embeddings = WordEmbeddingsModel().pretrained("embeddings_clinical", "en", "clinical/models")\
.setInputCols(["sentence", "token"]) \
.setOutputCol("embeddings")
ner = MedicalNerModel.pretrained("ner_oncology_wip", "en", "clinical/models") \
.setInputCols(["sentence", "token", "embeddings"]) \
.setOutputCol("ner")
ner_converter = NerConverter() \
.setInputCols(["sentence", "token", "ner"]) \
.setOutputCol("ner_chunk")\
.setWhiteList(["Cancer_Dx"])
assertion = AssertionDLModel.pretrained("assertion_oncology_problem_wip", "en", "clinical/models") \
.setInputCols(["sentence", "ner_chunk", "embeddings"]) \
.setOutputCol("assertion")
pipeline = Pipeline(stages=[document_assembler,
sentence_detector,
tokenizer,
word_embeddings,
ner,
ner_converter,
assertion])
data = spark.createDataFrame([["""The patient was diagnosed with breast cancer. Her family history is positive for other cancers."""]]).toDF("text")
result = pipeline.fit(data).transform(data)
val document_assembler = new DocumentAssembler()
.setInputCol("text")
.setOutputCol("document")
val sentence_detector = SentenceDetectorDLModel.pretrained("sentence_detector_dl_healthcare","en","clinical/models")
.setInputCols(Array("document"))
.setOutputCol("sentence")
val tokenizer = new Tokenizer()
.setInputCols(Array("sentence"))
.setOutputCol("token")
val word_embeddings = WordEmbeddingsModel().pretrained("embeddings_clinical", "en", "clinical/models")
.setInputCols(Array("sentence", "token"))
.setOutputCol("embeddings")
val ner = MedicalNerModel.pretrained("ner_oncology_wip", "en", "clinical/models")
.setInputCols(Array("sentence", "token", "embeddings"))
.setOutputCol("ner")
val ner_converter = new NerConverter()
.setInputCols(Array("sentence", "token", "ner"))
.setOutputCol("ner_chunk")
.setWhiteList(Array("Cancer_Dx"))
val clinical_assertion = AssertionDLModel.pretrained("assertion_oncology_problem_wip","en","clinical/models")
.setInputCols(Array("sentence","ner_chunk","embeddings"))
.setOutputCol("assertion")
val pipeline = new Pipeline().setStages(Array(document_assembler,
sentence_detector,
tokenizer,
word_embeddings,
ner,
ner_converter,
assertion))
val data = Seq("""The patient was diagnosed with breast cancer. Her family history is positive for other cancers.""").toDF("text")
val result = pipeline.fit(data).transform(data)
import nlu
nlu.load("en.assert.oncology_problem_wip").predict("""The patient was diagnosed with breast cancer. Her family history is positive for other cancers.""")
Results
| chunk | ner_label | assertion |
|:--------------|:------------|:----------------|
| breast cancer | Cancer_Dx | Medical_History |
| cancers | Cancer_Dx | Family_History |
Model Information
| Model Name: | assertion_oncology_problem_wip |
| Compatibility: | Healthcare NLP 4.1.0+ |
| License: | Licensed |
| Edition: | Official |
| Input Labels: | [document, chunk, embeddings] |
| Output Labels: | [assertion_pred] |
| Language: | en |
| Size: | 1.4 MB |
References
In-house annotated oncology case reports.
Benchmarking
label precision recall f1-score support
Absent 0.88 0.87 0.87 154.0
Family 0.67 1.00 0.80 8.0
Hypothetical 0.81 0.77 0.79 77.0
Possible 0.62 0.61 0.62 54.0
Present 0.78 0.79 0.78 155.0
macro-avg 0.75 0.81 0.77 448.0
weighted-avg 0.80 0.79 0.79 448.0