65 Medical text understanding PR

.github/workflows/main.yml

@@ -13,25 +13,17 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: [ 3.6, 3.7, 3.8, 3.9 ]
+        python-version: [ 3.7, 3.8, 3.9 ]
         torch-version: [ 1.5.0, 1.6.0, 1.7.1, 1.8.1 ]
         tensorflow-version: [ 1.15.0, 2.2.0, 2.5.0 ]
         test-details:
           - { dep: forte-wrapper, testfile: tests/fortex/health/processors/icd_coding_processor_test.py }
           - { dep: forte-wrapper, testfile: tests/fortex/health/processors/negation_context_analysis_test.py }
           - { dep: scispacy, testfile: tests/fortex/health/processors/scispacy_processor_test.py }
           - { testfile: tests/fortex/health/readers/mimic3_note_reader_test.py }
-          - { testfile: tests/fortex/health/readers/xray_image_reader_test.py }
-          - { testfile: tests/fortex/health/processors/xray_processor_test.py }
+          - { dep: xray_image_processor,testfile: tests/fortex/health/readers/xray_image_reader_test.py }
+          - { dep: xray_image_processor,testfile: tests/fortex/health/processors/xray_processor_test.py }
         exclude:
-          - python-version: 3.6
-            torch-version: 1.7.1
-          - python-version: 3.6
-            torch-version: 1.8.1
-          - python-version: 3.6
-            tensorflow-version: 2.2.0
-          - python-version: 3.6
-            tensorflow-version: 2.5.0
           - python-version: 3.7
             torch-version: 1.7.1
           - python-version: 3.7
@@ -72,7 +64,6 @@ jobs:
         restore-keys: |
           ${{ runner.os }}-pip-
           ${{ runner.os }}-
-
     - name: Install dependencies
       run: |
         sudo apt-get install -y libsndfile1-dev
@@ -81,49 +72,47 @@ jobs:
         pip install --progress-bar off pylint==2.10.2 importlib-metadata==4.8.0 flake8==3.9.2 mypy==0.931 pytest==5.1.3 black==20.8b1 click==8.0.1
         pip install --progress-bar off types-PyYAML==5.4.8 types-typed-ast==1.4.4 types-requests==2.25.6 types-dataclasses==0.1.7
         pip install --progress-bar off coverage codecov
-
     - name: Install deep learning frameworks
       run: |
         pip install --progress-bar off torch==${{ matrix.torch-version }}
         pip install --progress-bar off tensorflow==${{ matrix.tensorflow-version }}
-
     - name: Format check with Black
       run: |
         black --line-length 80 --check fortex/ tests/
-
     - name: Lint with flake8
       run: |
         # stop the build if there are Python syntax errors or undefined names
         flake8 fortex/ examples/ tests/ ftx/ --count --select=E9,F63,F7,F82 --show-source --statistics
         # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide
         flake8 fortex/ examples/ ftx/ tests/ --ignore E203,W503 --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
-
     - name: Lint with pylint
       run: |
         pylint fortex/ tests/
-
     - name: Lint main code with mypy when torch version is not 1.5.0 and python is 3.9
       run: |
         if [[ ${{ matrix.torch-version }} != "1.5.0" && ${{ matrix.python-version }} == "3.9" ]]; then mypy fortex/ tests/ ; fi
-
     - name: Install requirements for tests
       run: |
         pip install --progress-bar off .[test]
-
     - name: Install Forte-wrappers-spacy
       if: ${{ matrix.test-details.dep == 'forte-wrapper' ||
               contains(matrix.test-details.dep, 'forte-wrapper') }}
       run: |        
         git clone https://github.com/asyml/forte-wrappers.git
         cd forte-wrappers
         pip install src/spacy
-
     - name: Install scispacy
       if: ${{ matrix.test-details.dep == 'scispacy' ||
               contains(matrix.test-details.dep, 'scispacy') }}
       run: |        
         pip install --progress-bar off .[scispacy_processor]
-
+
+    - name: Install xray_image_processor
+      if: ${{ matrix.test-details.dep == 'xray_image_processor' ||
+              contains(matrix.test-details.dep, 'xray_image_processor') }}
+      run: |        
+        pip install --progress-bar off .[xray_image_processor]
+
     - name: Test with pytest and run coverage
       run: |
         coverage run -m pytest ${{ matrix.test-details.testfile}}
@@ -152,15 +141,13 @@ jobs:
           restore-keys: |
             ${{ runner.os }}-pip-
             ${{ runner.os }}-
-
       - name: Install dependencies
         run: |
           python -m pip install --progress-bar off --upgrade pip
           sudo apt-get install pandoc
           pip install --progress-bar off -r requirements.txt
           pip install --progress-bar off -r docs/requirements.txt
           pip install --progress-bar off .
-
       - name: Build Docs
         run: |
           cd docs

examples/medical_text_understanding/medical_text_understanding.py

@@ -0,0 +1,74 @@
+import sys
+from termcolor import colored
+
+from forte.data.data_pack import DataPack
+from forte.data.readers import PlainTextReader
+from forte.pipeline import Pipeline
+from forte.processors.writers import PackIdJsonPackWriter
+
+from ft.onto.base_ontology import (
+    Token,
+)
+
+from ftx.medical.clinical_ontology import Hyponym, Abbreviation
+from fortex.health.processors.scispacy_processor import (
+    ScispaCyProcessor,
+)
+
+def main(
+    input_path: str,
+    process: str,
+):
+    pl = Pipeline[DataPack]()
+    pl.set_reader(PlainTextReader())
+    if process == "abb":
+        pl.add(
+            ScispaCyProcessor(),
+            {
+                "entry_type": "ft.onto.base_ontology.Document",
+                "model_name": "en_core_sci_sm",
+                "pipe_name": "abbreviation_detector",
+                "prefer_gpu": True,
+                "require_gpu": False,
+                "gpu_id": 0,
+            },
+        )
+    elif process == "hyp":
+        pl.add(
+            ScispaCyProcessor(),
+            {
+                "entry_type": "ft.onto.base_ontology.Document",
+                "model_name": "en_core_sci_sm",
+                "pipe_name": "hyponym_detector",
+                "prefer_gpu": True,
+                "require_gpu": False,
+                "gpu_id": 0,
+            },
+        )
+
+    pl.initialize()
+
+    packs = pl.process_dataset(input_path)
+
+    for pack in packs:
+        if process == "abb":
+            print(colored("Input Sentence:", "blue"), colored( pack.text, "red"))
+            for idx, abv_item in enumerate(pack.get(Abbreviation)):
+                print(colored("Abbreviation:", "blue"), colored( abv_item.text, "green"))
+                print(colored("Long form text:", "blue"), colored( abv_item.long_form.text, "green"), "\n")
+
+        elif process == "hyp":
+            print(colored("Input Sentence:", "blue"), colored( pack.text, "red"))
+            for idx, detected in enumerate(pack.get(Hyponym)):
+                print(colored("hyponym_link:", "blue"), colored(detected.hyponym_link, "green"))
+                print(colored("hypernym:", "blue"), colored(detected.general, "green"))
+                print(colored("hyponym:", "blue"), colored(detected.specific, "green"))
+# Examples:
+'''
+Read from sample_data:
+Abbreviation:
+    python medical_text_understanding.py sample_data_abb/ abb
+Hyponym detection:
+    python medical_text_understanding.py sample_data_hyp/ hyp
+'''
+main(sys.argv[1], sys.argv[2])

examples/medical_text_understanding/sample_data_abb/abb.txt

@@ -0,0 +1,2 @@
+Spinal and bulbar muscular atrophy (SBMA) is an inherited motor neuron disease caused by the expansion of a polyglutamine tract within the androgen 
+receptor (AR). SBMA can be caused by this easily.

examples/medical_text_understanding/sample_data_hyp/hyp.txt

@@ -0,0 +1 @@
+Keystone plant species such as fig trees are good for the soil.

fortex/health/readers/xray_image_reader.py

@@ -26,7 +26,7 @@
 
 
 class XrayImageReader(PackReader):
-    r""":class:`ImageReader` is designed to read image files from a given folder."""
+    r""":class:`XrayImageReader` is designed to read image files from a given folder."""
 
     def __init__(self):
         super().__init__()
@@ -64,7 +64,7 @@ def default_configs(cls):
 
         Here:
 
-          - file_ext (str): The file extension to find the target audio files
+          - file_ext (str): The file extension to find the target files
              under a specific directory path. Default value is ".jpeg".
 
           - read_kwargs (dict): A dictionary containing all the keyword

setup.py

@@ -28,15 +28,19 @@
         "test": [
             "ddt",
             "testfixtures",
-            "transformers==4.18.0",
+            "transformers==4.2.2",
             "protobuf==3.19.4",
-            "Pillow==8.4.0",
+            "numpy==1.21.6",
             'forte @ git+https://github.com/asyml/forte',
         ],
         "scispacy_processor": [
             "scispacy==0.5.0",
             "en-core-sci-sm @ https://s3-us-west-2.amazonaws.com/ai2-s2-scispacy/releases/v0.5.0/en_core_sci_sm-0.5.0.tar.gz"
         ],
+        "xray_image_processor": [
+            "Pillow==8.4.0",
+            "transformers==4.18.0",
+        ],
     },
     include_package_data=True,
     python_requires='>=3.6',