Description
This model detects the assertion status of oncology tests, such as Pathology_Test or Imaging_Test. The model identifies if the test has been performed (Present_Or_Past status), or if it is mentioned but in hypothetical or absent sentences (Hypothetical_Or_Absent status).
Predicted Entities
Hypothetical_Or_Absent, Medical_History
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(["Pathology_Test", "Imaging_Test"])
assertion = AssertionDLModel.pretrained("assertion_oncology_test_binary", "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 result of the biopsy was positive. We recommend to perform a CT scan."]]).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("Pathology_Test", "Imaging_Test"))
val clinical_assertion = AssertionDLModel.pretrained("assertion_oncology_test_binary","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 result of the biopsy was positive. We recommend to perform a CT scan.""").toDF("text")
val result = pipeline.fit(data).transform(data)
Results
| chunk | ner_label | assertion |
|:--------|:---------------|:-----------------------|
| biopsy | Pathology_Test | Medical_History |
| CT scan | Imaging_Test | Hypothetical_Or_Absent |
Model Information
| Model Name: | assertion_oncology_test_binary |
| Compatibility: | Healthcare NLP 5.4.0+ |
| 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
Hypothetical_Or_Absent 0.79 0.81 0.80 37.0
Medical_History 0.80 0.78 0.79 36.0
macro-avg 0.79 0.79 0.79 73.0
weighted-avg 0.79 0.79 0.79 73.0