SDOH Under Treatment For Classification

document_assembler = DocumentAssembler()\
    .setInputCol("text")\
    .setOutputCol("document")
        
sentence_embeddings = BertSentenceEmbeddings.pretrained("sbiobert_base_cased_mli", 'en','clinical/models')\
    .setInputCols(["document"])\
    .setOutputCol("sentence_embeddings")

features_asm = FeaturesAssembler()\
    .setInputCols(["sentence_embeddings"])\
    .setOutputCol("features")

generic_classifier = GenericClassifierModel.pretrained("genericclassifier_sdoh_under_treatment_sbiobert_cased_mli", 'en', 'clinical/models')\
    .setInputCols(["features"])\
    .setOutputCol("prediction")

pipeline = Pipeline(stages=[
    document_assembler,
    sentence_embeddings,
    features_asm,
    generic_classifier    
])

text_list = ["Patient is a 50-year-old male who was diagnosed with hepatitis C. He has received a treatment plan that includes medication and regular monitoring of his liver function.",
                 
"Patient has been living with chronic migraines for several years. She has not pursued any specific treatment for her migraines and has been managing her condition through lifestyle modifications such as stress reduction techniques and avoiding triggers.",
                
"""Sarah, a 55-year-old woman with a history of high cholesterol and a family history of heart disease, presented to her primary care physician with complaints of chest pain and shortness of breath. After a thorough evaluation, Sarah was diagnosed with coronary artery disease (CAD), a condition that can lead to heart attacks and other serious complications.

To manage her CAD, Sarah was started on a treatment plan that included medication to lower her cholesterol and blood pressure, as well as aspirin to prevent blood clots. In addition to medication, Sarah was advised to make lifestyle modifications such as improving her diet, quitting smoking, and increasing physical activity.

Over the course of several months, Sarah's symptoms improved, and follow-up tests showed that her cholesterol and blood pressure were within the target range. However, Sarah continued to experience occasional chest pain, and her medication regimen was adjusted accordingly.

With regular follow-up appointments and adherence to her treatment plan, Sarah's CAD remained under control, and she was able to resume her normal activities with improved quality of life.
""",
    
"""John, a 60-year-old man with a history of smoking and high blood pressure, presented to his primary care physician with complaints of chest pain and shortness of breath. Further tests revealed that John had a blockage in one of his coronary arteries, which required urgent intervention. However, John was hesitant to undergo treatment, citing concerns about potential complications and side effects of medications and procedures.

Despite the physician's recommendations and attempts to educate John about the risks of leaving the blockage untreated, John ultimately chose not to pursue any treatment. Over the next several months, John continued to experience symptoms, which progressively worsened, and he ultimately required hospitalization for a heart attack. The medical team attempted to intervene at that point, but the damage to John's heart was severe, and his prognosis was poor.
"""]


df = spark.createDataFrame(text_list, StringType()).toDF("text")

result = pipeline.fit(df).transform(df)

result.select("text", "prediction.result").show(truncate=100)