Update app.py
Browse files
app.py
CHANGED
@@ -18,11 +18,17 @@ import os
|
|
18 |
|
19 |
# Load terminology datasets:
|
20 |
basedir = os.path.dirname(__file__)
|
21 |
-
dataLOINC = pd.read_csv(basedir + f'LoincTableCore.csv')
|
22 |
-
dataPanels = pd.read_csv(basedir + f'PanelsAndForms-ACW1208Labeled.csv')
|
23 |
-
dataSNOMED = pd.read_csv(basedir + f'sct2_TextDefinition_Full-en_US1000124_20220901.txt',sep='\t')
|
24 |
-
dataOMS = pd.read_csv(basedir + f'SnomedOMS.csv')
|
25 |
-
dataICD10 = pd.read_csv(basedir + f'ICD10Diagnosis.csv')
|
|
|
|
|
|
|
|
|
|
|
|
|
26 |
|
27 |
dir_path = os.path.dirname(os.path.realpath(__file__))
|
28 |
EXAMPLES = {}
|
@@ -168,8 +174,8 @@ def group_by_entity(raw):
|
|
168 |
if g21 != 'Series([] )': SaveResult(s2, outputFile)
|
169 |
|
170 |
#SNOMED
|
171 |
-
g31 = g3['conceptId'].to_string().replace(","," ").replace("\n"," ")
|
172 |
-
g32 = g3['term'].to_string().replace(","," ").replace("\n"," ")
|
173 |
s3 = ("SNOMED Terms of entity ," + myEntityGroup + ", with term ," + eterm + ", SNOMED concepts of ," + g31 + ", and SNOMED terms of ," + g32 + ", Label,Value, Label,Value, Label,Value ")
|
174 |
if g31 != 'Series([] )': SaveResult(s3, outputFile)
|
175 |
|
@@ -240,7 +246,7 @@ def ner(text):
|
|
240 |
|
241 |
#return (ner_content, meta, label, figure)
|
242 |
outputDataframe = pd.read_csv(outputFile)
|
243 |
-
outputFile = outputFile.replace(os.path.dirname(__file__) + "\\","") # Just filename for File download UI output element
|
244 |
|
245 |
#return (ner_content, meta, label, figure, outputDataframe, outputFile)
|
246 |
return (ner_content, outputDataframe, outputFile)
|
@@ -250,13 +256,13 @@ demo = gr.Blocks()
|
|
250 |
with demo:
|
251 |
gr.Markdown(
|
252 |
"""
|
253 |
-
# 🩺⚕️NLP
|
254 |
"""
|
255 |
)
|
256 |
input = gr.Textbox(label="Note text", value="")
|
257 |
-
output=[
|
258 |
-
|
259 |
-
]
|
260 |
with gr.Tab("Biomedical Entity Recognition"):
|
261 |
output=[
|
262 |
gr.HighlightedText(label="NER", combine_adjacent=True),
|
@@ -280,12 +286,12 @@ with demo:
|
|
280 |
btnOMS = gr.Button("OMS")
|
281 |
btnICD10 = gr.Button("ICD10")
|
282 |
|
283 |
-
output=[
|
284 |
-
|
285 |
-
|
286 |
-
|
287 |
-
|
288 |
-
]
|
289 |
|
290 |
|
291 |
#textCT = gr.Textbox(placeholder="CT Match Results", lines=10)
|
|
|
18 |
|
19 |
# Load terminology datasets:
|
20 |
basedir = os.path.dirname(__file__)
|
21 |
+
#dataLOINC = pd.read_csv(basedir + "\\" + f'LoincTableCore.csv')
|
22 |
+
#dataPanels = pd.read_csv(basedir + "\\" + f'PanelsAndForms-ACW1208Labeled.csv')
|
23 |
+
#dataSNOMED = pd.read_csv(basedir + "\\" + f'sct2_TextDefinition_Full-en_US1000124_20220901.txt',sep='\t')
|
24 |
+
#dataOMS = pd.read_csv(basedir + "\\" + f'SnomedOMS.csv')
|
25 |
+
#dataICD10 = pd.read_csv(basedir + "\\" + f'ICD10Diagnosis.csv')
|
26 |
+
|
27 |
+
dataLOINC = pd.read_csv(f'LoincTableCore.csv')
|
28 |
+
dataPanels = pd.read_csv(f'PanelsAndForms-ACW1208Labeled.csv')
|
29 |
+
dataSNOMED = pd.read_csv(f'sct2_TextDefinition_Full-en_US1000124_20220901.txt',sep='\t')
|
30 |
+
dataOMS = pd.read_csv(f'SnomedOMS.csv')
|
31 |
+
dataICD10 = pd.read_csv(f'ICD10Diagnosis.csv')
|
32 |
|
33 |
dir_path = os.path.dirname(os.path.realpath(__file__))
|
34 |
EXAMPLES = {}
|
|
|
174 |
if g21 != 'Series([] )': SaveResult(s2, outputFile)
|
175 |
|
176 |
#SNOMED
|
177 |
+
g31 = g3['conceptId'].to_string().replace(","," ").replace("\n"," ").replace("\l"," ").replace("\r"," ")
|
178 |
+
g32 = g3['term'].to_string().replace(","," ").replace("\n"," ").replace("\l"," ").replace("\r"," ")
|
179 |
s3 = ("SNOMED Terms of entity ," + myEntityGroup + ", with term ," + eterm + ", SNOMED concepts of ," + g31 + ", and SNOMED terms of ," + g32 + ", Label,Value, Label,Value, Label,Value ")
|
180 |
if g31 != 'Series([] )': SaveResult(s3, outputFile)
|
181 |
|
|
|
246 |
|
247 |
#return (ner_content, meta, label, figure)
|
248 |
outputDataframe = pd.read_csv(outputFile)
|
249 |
+
#outputFile = outputFile.replace(os.path.dirname(__file__) + "\\","") # Just filename for File download UI output element
|
250 |
|
251 |
#return (ner_content, meta, label, figure, outputDataframe, outputFile)
|
252 |
return (ner_content, outputDataframe, outputFile)
|
|
|
256 |
with demo:
|
257 |
gr.Markdown(
|
258 |
"""
|
259 |
+
# 🩺⚕️NLP Clinical Ontology Biomedical NER Model
|
260 |
"""
|
261 |
)
|
262 |
input = gr.Textbox(label="Note text", value="")
|
263 |
+
#output=[
|
264 |
+
# gr.HighlightedText(label="NER", combine_adjacent=True)
|
265 |
+
#]
|
266 |
with gr.Tab("Biomedical Entity Recognition"):
|
267 |
output=[
|
268 |
gr.HighlightedText(label="NER", combine_adjacent=True),
|
|
|
286 |
btnOMS = gr.Button("OMS")
|
287 |
btnICD10 = gr.Button("ICD10")
|
288 |
|
289 |
+
#output=[
|
290 |
+
# gr.HighlightedText(label="NER", combine_adjacent=True),
|
291 |
+
# gr.File(label="File"), # add download link here
|
292 |
+
# gr.Dataframe(label="Dataframe", headers=["LOINC", "Panels", "SNOMED", "OMS", "ICD10"]), # add harmonised output for input corpus here as a dataframe to UI
|
293 |
+
# gr.Textbox(placeholder="CT Match Results", lines=10) # add matched text scratchpad here
|
294 |
+
#]
|
295 |
|
296 |
|
297 |
#textCT = gr.Textbox(placeholder="CT Match Results", lines=10)
|