File size: 11,526 Bytes
157ac27
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
import gradio as gr
import pandas as pd
import json
from collections import defaultdict

# Create tokenizer for biomed model
from transformers import pipeline, AutoTokenizer, AutoModelForTokenClassification
tokenizer = AutoTokenizer.from_pretrained("d4data/biomedical-ner-all")    # https://huggingface.co/d4data/biomedical-ner-all?text=asthma
model = AutoModelForTokenClassification.from_pretrained("d4data/biomedical-ner-all")
pipe = pipeline("ner", model=model, tokenizer=tokenizer, aggregation_strategy="simple")

# Matplotlib for entity graph
import matplotlib.pyplot as plt
plt.switch_backend("Agg")

# Load examples from JSON
import os 

# Load terminology datasets:
basedir = os.path.dirname(__file__)
#dataLOINC = pd.read_csv(basedir + "\\" + f'LoincTableCore.csv')
#dataPanels = pd.read_csv(basedir + "\\" + f'PanelsAndForms-ACW1208Labeled.csv')     
#dataSNOMED = pd.read_csv(basedir + "\\" + f'sct2_TextDefinition_Full-en_US1000124_20220901.txt',sep='\t')   
#dataOMS = pd.read_csv(basedir + "\\" + f'SnomedOMS.csv')   
#dataICD10 = pd.read_csv(basedir + "\\" + f'ICD10Diagnosis.csv') 

dataLOINC = pd.read_csv(f'LoincTableCore.csv')
dataPanels = pd.read_csv(f'PanelsAndForms-ACW1208Labeled.csv')     
dataSNOMED = pd.read_csv(f'sct2_TextDefinition_Full-en_US1000124_20220901.txt',sep='\t')   
dataOMS = pd.read_csv(f'SnomedOMS.csv')   
dataICD10 = pd.read_csv(f'ICD10Diagnosis.csv')   

dir_path = os.path.dirname(os.path.realpath(__file__))
EXAMPLES = {}
#with open(dir_path + "\\" + "examples.json", "r") as f:
with open("examples.json", "r") as f:
    example_json = json.load(f)
    EXAMPLES = {x["text"]: x["label"] for x in example_json}

def MatchLOINC(name):
    #basedir = os.path.dirname(__file__)
    pd.set_option("display.max_rows", None)
    #data = pd.read_csv(basedir + "\\" + f'LoincTableCore.csv')    
    data = dataLOINC
    swith=data.loc[data['COMPONENT'].str.contains(name, case=False, na=False)]
    return swith
    
def MatchLOINCPanelsandForms(name):
    #basedir = os.path.dirname(__file__)
    #data = pd.read_csv(basedir + "\\" + f'PanelsAndForms-ACW1208Labeled.csv')     
    data = dataPanels
    # Assessment Name:
    #swith=data.loc[data['ParentName'].str.contains(name, case=False, na=False)]
    # Assessment Question:
    swith=data.loc[data['LoincName'].str.contains(name, case=False, na=False)]
    return swith
    
def MatchSNOMED(name):
    #basedir = os.path.dirname(__file__)
    #data = pd.read_csv(basedir + "\\" + f'sct2_TextDefinition_Full-en_US1000124_20220901.txt',sep='\t')   
    data = dataSNOMED
    swith=data.loc[data['term'].str.contains(name, case=False, na=False)]
    return swith

def MatchOMS(name):
    #basedir = os.path.dirname(__file__)
    #data = pd.read_csv(basedir + "\\" + f'SnomedOMS.csv')   
    data = dataOMS
    swith=data.loc[data['SNOMED CT'].str.contains(name, case=False, na=False)]
    return swith

def MatchICD10(name):
    #basedir = os.path.dirname(__file__)
    #data = pd.read_csv(basedir + "\\" + f'ICD10Diagnosis.csv')   
    data = dataICD10
    swith=data.loc[data['Description'].str.contains(name, case=False, na=False)]
    return swith

def SaveResult(text, outputfileName):
    #try:
    basedir = os.path.dirname(__file__)
    savePath = outputfileName
    print("Saving: " + text + " to " + savePath)
    from os.path import exists
    file_exists = exists(savePath)
    if file_exists:
        with open(outputfileName, "a") as f: #append
            #for line in text:
            f.write(str(text.replace("\n","  ")))
            f.write('\n')
    else:
        with open(outputfileName, "w") as f: #write
            #for line in text:
            f.write(str(text.replace("\n","  ")))
            f.write('\n')
    #except ValueError as err:
    #    raise ValueError("File Save Error in SaveResult \n" + format_tb(err.__traceback__)[0] + err.args[0] + "\nEnd of error message.") from None

    return

def loadFile(filename):
    try:
        basedir = os.path.dirname(__file__)
        loadPath = basedir + "\\" + filename

        print("Loading: " + loadPath)

        from os.path import exists
        file_exists = exists(loadPath)

        if file_exists:
            with open(loadPath, "r") as f: #read
                contents = f.read()
                print(contents)
                return contents

    except ValueError as err:
        raise ValueError("File Save Error in SaveResult \n" + format_tb(err.__traceback__)[0] + err.args[0] + "\nEnd of error message.") from None

    return ""

def get_today_filename():
    from datetime import datetime
    date = datetime.now().strftime("%Y_%m_%d-%I.%M.%S.%p")
    #print(f"filename_{date}")  'filename_2023_01_12-03-29-22_AM'
    return f"MedNER_{date}.csv"

def get_base(filename): 
        basedir = os.path.dirname(__file__)
        loadPath = basedir + "\\" + filename
        #print("Loading: " + loadPath)
        return loadPath

def group_by_entity(raw):
    outputFile = get_base(get_today_filename())
    out = defaultdict(int)

    for ent in raw:
        out[ent["entity_group"]] += 1
        myEntityGroup = ent["entity_group"]
        print("Found entity group type: " + myEntityGroup)

        if (myEntityGroup in ['Sign_symptom', 'Detailed_description', 'History', 'Activity', 'Medication' ]):
            eterm = ent["word"].replace('#','')
            minlength = 3
            if len(eterm) > minlength:
                print("Found eterm: " + eterm)
                eterm.replace("#","")
                g1=MatchLOINC(eterm)
                g2=MatchLOINCPanelsandForms(eterm)
                g3=MatchSNOMED(eterm)
                g4=MatchOMS(eterm)
                g5=MatchICD10(eterm)
                sAll = ""

                print("Saving to output file " + outputFile)
                # Create harmonisation output format of input to output code, name, Text

                try: # 18 fields, output to labeled CSV dataset for results teaching on scored regret changes to action plan with data inputs
                    col = "          1                            2            3         4            5                     6                    7                       8                   9              10                   11                         12       13               14                      15                  16                            17                    18                       19"
                    
                    #LOINC
                    g11 = g1['LOINC_NUM'].to_string().replace(","," ").replace("\n"," ")
                    g12 = g1['COMPONENT'].to_string().replace(","," ").replace("\n"," ")
                    s1 = ("LOINC," + myEntityGroup + "," + eterm + ",questions of ," + g12 + "," + g11 + ", Label,Value, Label,Value, Label,Value  ")
                    if g11 != 'Series([]  )': SaveResult(s1, outputFile)

                    #LOINC Panels
                    g21 = g2['Loinc'].to_string().replace(","," ").replace("\n"," ")
                    g22 = g2['LoincName'].to_string().replace(","," ").replace("\n"," ")
                    g23 = g2['ParentLoinc'].to_string().replace(","," ").replace("\n"," ")
                    g24 = g2['ParentName'].to_string().replace(","," ").replace("\n"," ")
                    # s2 = ("LOINC Panel," + myEntityGroup + "," + eterm + ",name of ," + g22 + "," + g21 + ", and Parent codes of ," + g23 + ", with Parent names of ," + g24 + ", Label,Value  ")
                    s2 = ("LOINC Panel," + myEntityGroup + "," + eterm + ",name of ," + g22 + "," + g21 + "," + g24 + ", and Parent codes of ," + g23 + "," + ", Label,Value  ")
                    if g21 != 'Series([]  )': SaveResult(s2, outputFile)

                    #SNOMED
                    g31 = g3['conceptId'].to_string().replace(","," ").replace("\n"," ").replace("\l"," ").replace("\r"," ")
                    g32 = g3['term'].to_string().replace(","," ").replace("\n"," ").replace("\l"," ").replace("\r"," ")
                    s3 = ("SNOMED Concept," + myEntityGroup + "," + eterm + ",terms of ," + g32 + "," + g31 + ", Label,Value, Label,Value, Label,Value  ")
                    if g31 != 'Series([]  )': SaveResult(s3, outputFile)

                    #OMS
                    g41 = g4['Omaha Code'].to_string().replace(","," ").replace("\n"," ")
                    g42 = g4['SNOMED CT concept ID'].to_string().replace(","," ").replace("\n"," ")
                    g43 = g4['SNOMED CT'].to_string().replace(","," ").replace("\n"," ")
                    g44 = g4['PR'].to_string().replace(","," ").replace("\n"," ")
                    g45 = g4['S&S'].to_string().replace(","," ").replace("\n"," ")
                    s4 = ("OMS," + myEntityGroup + "," + eterm + ",concepts of ," + g44 + "," + g45 + ", and SNOMED codes of ," + g43 + ", and OMS problem of ," + g42 + ", and OMS Sign Symptom of ," + g41)
                    if g41 != 'Series([]  )': SaveResult(s4, outputFile)

                    #ICD10
                    g51 = g5['Code'].to_string().replace(","," ").replace("\n"," ")
                    g52 = g5['Description'].to_string().replace(","," ").replace("\n"," ")
                    s5 = ("ICD10," + myEntityGroup + "," + eterm + ",descriptions of ," + g52 + "," + g51 + ", Label,Value, Label,Value, Label,Value  ")
                    if g51 != 'Series([]  )': SaveResult(s5, outputFile)

                except ValueError as err:
                    raise ValueError("Error in group by entity \n" + format_tb(err.__traceback__)[0] + err.args[0] + "\nEnd of error message.") from None

    return outputFile


def plot_to_figure(grouped):
    fig = plt.figure()
    plt.bar(x=list(grouped.keys()), height=list(grouped.values()))
    plt.margins(0.2)
    plt.subplots_adjust(bottom=0.4)
    plt.xticks(rotation=90)
    return fig


def ner(text):
    raw = pipe(text)
    ner_content = {
        "text": text,
        "entities": [
            {
                "entity": x["entity_group"],
                "word": x["word"],
                "score": x["score"],
                "start": x["start"],
                "end": x["end"],
            }
            for x in raw
        ],
    }
    
    outputFile = group_by_entity(raw)
    label = EXAMPLES.get(text, "Unknown")
    outputDataframe = pd.read_csv(outputFile)
    return (ner_content, outputDataframe, outputFile)

demo = gr.Blocks()
with demo:
    gr.Markdown(
    """
    # 🩺⚕️NLP Clinical Ontology Biomedical NER 
    """
    )
    input = gr.Textbox(label="Note text", value="")

    with gr.Tab("Biomedical Entity Recognition"):
        output=[
            gr.HighlightedText(label="NER", combine_adjacent=True),
            #gr.JSON(label="Entity Counts"),
            #gr.Label(label="Rating"),
            #gr.Plot(label="Bar"),
            gr.Dataframe(label="Dataframe"),
            gr.File(label="File"),
        ]
        examples=list(EXAMPLES.keys())  
        gr.Examples(examples, inputs=input)
        input.change(fn=ner, inputs=input, outputs=output)
        
    with gr.Tab("Clinical Terminology Resolution"):
        with gr.Row(variant="compact"):
            btnLOINC = gr.Button("LOINC")
            btnPanels = gr.Button("Panels")
            btnSNOMED = gr.Button("SNOMED")
            btnOMS = gr.Button("OMS")
            btnICD10 = gr.Button("ICD10")

        examples=list(EXAMPLES.keys())  
        gr.Examples(examples, inputs=input)
        input.change(fn=ner, inputs=input, outputs=output)
#layout="vertical"
demo.launch(debug=True)