Description
This model extracts cancer therapies (Cancer_Surgery, Radiotherapy, and Cancer_Therapy) and posology information at a granular level. It is the version of ner_oncology_posology model augmented with langtest library.
Definitions of Predicted Entities:
Cancer_Surgery: Terms that indicate surgery as a form of cancer treatment.Cancer_Therapy: Any cancer treatment mentioned in text, excluding surgeries and radiotherapy.Cycle_Count: The total number of cycles being administered of an oncological therapy (e.g. “5 cycles”).Cycle_Day: References to the day of the cycle of oncological therapy (e.g. “day 5”).Cycle_Number: The number of the cycle of an oncological therapy that is being applied (e.g. “third cycle”).Dosage: The quantity prescribed by the physician for an active ingredient.Duration: Words indicating the duration of a treatment (e.g. “for 2 weeks”).Frequency: Words indicating the frequency of treatment administration (e.g. “daily” or “bid”).Radiotherapy: Terms that indicate the use of Radiotherapy.Radiation_Dose: Dose used in radiotherapy.Route: Words indicating the type of administration route (such as “PO” or “transdermal”).
Predicted Entities
Cancer_Surgery, Cancer_Therapy, Cycle_Count, Cycle_Day, Cycle_Number, Dosage, Duration, Frequency, Radiotherapy, Radiation_Dose, Route
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_posology_langtest", "en", "clinical/models") \
.setInputCols(["sentence", "token", "embeddings"]) \
.setOutputCol("ner")
ner_converter = NerConverter() \
.setInputCols(["sentence", "token", "ner"]) \
.setOutputCol("ner_chunk")
pipeline = Pipeline(stages=[document_assembler,
sentence_detector,
tokenizer,
word_embeddings,
ner,
ner_converter])
data = spark.createDataFrame([["The patient underwent a regimen consisting of adriamycin (60 mg/m2) and cyclophosphamide (600 mg/m2) over six courses. She is currently receiving his second cycle of chemotherapy and is in good overall condition."]]).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("document")
.setOutputCol("sentence")
val tokenizer = new Tokenizer()
.setInputCols("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_posology_langtest", "en", "clinical/models")
.setInputCols(Array("sentence", "token", "embeddings"))
.setOutputCol("ner")
val ner_converter = new NerConverter()
.setInputCols(Array("sentence", "token", "ner"))
.setOutputCol("ner_chunk")
val pipeline = new Pipeline().setStages(Array(
document_assembler,
sentence_detector,
tokenizer,
word_embeddings,
ner,
ner_converter))
val data = Seq("The patient underwent a regimen consisting of adriamycin (60 mg/m2) and cyclophosphamide (600 mg/m2) over six courses. She is currently receiving his second cycle of chemotherapy and is in good overall condition.").toDS.toDF("text")
val result = pipeline.fit(data).transform(data)
Results
+----------------+--------------+
|chunk |ner_label |
+----------------+--------------+
|adriamycin |Cancer_Therapy|
|60 mg/m2 |Dosage |
|cyclophosphamide|Cancer_Therapy|
|600 mg/m2 |Dosage |
|six courses |Cycle_Count |
|second cycle |Cycle_Number |
|chemotherapy |Cancer_Therapy|
+----------------+--------------+
Model Information
| Model Name: | ner_oncology_posology_langtest |
| Compatibility: | Healthcare NLP 5.0.2+ |
| License: | Licensed |
| Edition: | Official |
| Input Labels: | [sentence, token, embeddings] |
| Output Labels: | [ner] |
| Language: | en |
| Size: | 14.7 MB |
References
In-house annotated oncology case reports.
Benchmarking
label precision recall f1-score support
B-Cancer_Therapy 0.93 0.96 0.94 1185
B-Dosage 0.90 0.88 0.89 258
I-Dosage 0.90 0.94 0.92 752
B-Frequency 0.92 0.92 0.92 157
I-Frequency 0.92 0.86 0.89 218
B-Cancer_Surgery 0.83 0.85 0.84 517
I-Cancer_Therapy 0.81 0.86 0.83 507
B-Radiotherapy 0.91 0.86 0.89 170
I-Radiotherapy 0.91 0.75 0.82 120
B-Duration 0.87 0.79 0.83 280
I-Duration 0.89 0.85 0.87 537
I-Cancer_Surgery 0.75 0.78 0.77 370
B-Cycle_Number 0.89 0.61 0.72 41
I-Cycle_Number 0.89 0.61 0.72 41
B-Cycle_Count 0.82 0.87 0.84 128
I-Cycle_Count 0.83 0.88 0.86 115
I-Radiation_Dose 0.93 0.86 0.89 77
B-Cycle_Day 0.85 0.85 0.85 124
B-Route 0.91 0.92 0.92 114
I-Cycle_Day 0.87 0.77 0.82 177
I-Route 0.81 0.72 0.76 29
B-Radiation_Dose 0.93 0.95 0.94 43
micro-avg 0.88 0.88 0.88 5960
macro-avg 0.88 0.83 0.85 5960
weighted-avg 0.88 0.88 0.88 5960