File size: 13,422 Bytes
371da07
 
22b21fc
 
 
 
371da07
 
 
 
 
 
 
 
 
 
 
42e4a8b
 
371da07
42e4a8b
 
 
371da07
42e4a8b
 
 
371da07
42e4a8b
 
 
 
 
 
 
371da07
 
42e4a8b
 
 
 
 
371da07
42e4a8b
371da07
42e4a8b
 
 
 
 
 
 
 
 
 
371da07
42e4a8b
371da07
42e4a8b
 
 
 
 
 
 
 
 
 
 
 
 
 
371da07
 
 
 
 
 
42e4a8b
 
 
 
 
 
371da07
42e4a8b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
371da07
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
42e4a8b
371da07
 
 
 
 
 
 
 
 
 
 
 
 
42e4a8b
371da07
 
 
 
 
 
 
 
 
42e4a8b
371da07
 
 
42e4a8b
371da07
42e4a8b
371da07
 
42e4a8b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
371da07
 
 
 
 
 
 
42e4a8b
371da07
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
import json
from config import google_api
import os
import base64
from google import genai
from google.genai import types


def process_text(extracted_text):
    client = genai.Client(
        api_key=google_api,
    )
    model = "gemini-2.0-flash"
    contents = [
        types.Content(
            role="user",
            parts=[
                types.Part.from_text(text="""**Instruction:**
You are an advanced AI model specializing in medical data extraction. Given an unstructured OCR-extracted text from a medical lab report, your task is to:

1. **Correct Errors**
   - Fix missing decimals, incorrect test names, and incorrect reference ranges.
   - Ensure test values fall within valid medical reference ranges.

2. **Extract and Structure Data**
   - Extract **metadata** (patient details) and **lab report data** in structured JSON format.
   - Maintain consistency in naming conventions and JSON structure.

3. **Assign Status Labels**
   - **GREEN**: Value is within the normal range.
   - **AMBER**: Borderline or slightly out of range.
   - **RED**: Critical or significantly out of range.

### **JSON Output Format (Strictly Follow This Structure)**
```json
{
    \"metadata\": {
        \"patient_name\": \"<Corrected Name>\",
        \"age\": \"<Age>\",
        \"gender\": \"<Male/Female>\",
        \"lab_name\": \"<Lab Name>\",
        \"report_date\": \"<DD-MM-YYYY>\"
    },
    \"report\": [
        {
            \"test_type\": \"<HEMOGRAM / BIOCHEMISTRY / OTHER>\",
            \"lab_tests\": [
                {
                    \"test_name\": \"<Corrected Test Name>\",
                    \"value\": \"<Numerical Value>\",
                    \"unit\": \"<Unit>\",
                    \"reference_range\": \"<Lower Limit - Upper Limit Unit>\",
                    \"status\": \"<GREEN / AMBER / RED>\"
                }
            ]
        }
    ]
}
```
###EXTRACTED TEXT :
Dr. Onkar Test Sanjeevan Iospital MNES Mn) No:Tiz 12/4 Paud Racid  Kothrud Fune - 4V102 Ph: 02025262+5,8983390126, Tlmins: 09.15 AM 0z.30 PMOS.30PM OY_OPAAPPOINTMENTS ONLY | Closed: Mondjy Fridwy Ftent UID: 67 Report No: UOOI8 Nane: AMAF SHAHA (Mle) DIc 02-lul-20 73e 40 years Sample CollectedAc HoqitLb Mddress; MG Rozd FUNE Simple Type/Quantly: Blood Ref. By Doctor Sumnple Collexulon DT: 2-Jul-20, 950AV Dc . Amlt Dcshmukh Tesl Fesult DT: 0)-Jul-zo. +53PI HEMOGRAM IvesTGATiUR RESULT UNM REF. MINGE HaLMOGiOMN 14 guts/a 120.170 KRC coint 4 nlfcunm 41 51 HAEMOTOCRIT (PcW) 30 9u 320.470 MCV 78 n 760.i00, Mch H 32 6 200.320 McHc 32 A 315.365 Rdw 13 9 ; 116.I50 MPv 112 6A . 126 WKC CNT 4567 IcuMIM ACOO Tiuoo DIFTERENTML CnT MeuTKOHILS 56 10 - 70 LyMPOCTTES 20 200 450 NONcytis 1 4.6 RASOPHILS 0.1 PETELETS 273+00 fcu mm I50oou -+Ouuou PEIIPHERIL SMEAIL EXAMINATION RBC Mopholegy Nonmocyulc Normaxchronulc PMTELETS Adequate RS For "'Jnsites Not Scen TEST DONE ON FULLY AUTOMATED) FWVE PART HAEMOTOLOGY ANALYSER XTIROO , SYSMEXHPENTRA XLR HORRIRIBA (DfierentaalWBC Cunt [s correlted microscoplelly) H High, L Loiv ) END OF REPORT 8 Rv Rc




### **Key Constraints:**
- **DO NOT** include extra metadata fields beyond `patient_name`, `age`, `gender`, `lab_name`, and `report_date`.
- **DO NOT** include extra test fields beyond `test_name`, `value`, `unit`, `reference_range`, and `status`.
- **ENSURE** numerical values align with real-world medical reference ranges.
- **FORMAT OUTPUT** strictly in JSON, ensuring proper nesting.

"""),
            ],
        ),
        types.Content(
            role="model",
            parts=[
                types.Part.from_text(text="""{
    \"metadata\": {
        \"patient_name\": \"Amar Shaha\",
        \"age\": \"40 years\",
        \"gender\": \"Male\",
        \"lab_name\": \"Sanjeevan Hospital\",
        \"report_date\": \"02-07-2023\"
    },
    \"report\": [
        {
            \"lab_tests\": [
                {
                    \"reference_range\": \"12 - 17 g/dL\",
                    \"status\": \"GREEN\",
                    \"test_name\": \"Hemoglobin\",
                    \"unit\": \"g/dL\",
                    \"value\": \"14.1\"
                },
                {
                    \"reference_range\": \"4 - 6 million/cu.mm\",
                    \"status\": \"GREEN\",
                    \"test_name\": \"RBC Count\",
                    \"unit\": \"million/cu.mm\",
                    \"value\": \"4.8\"
                },
                {
                    \"reference_range\": \"37 - 52 %\",
                    \"status\": \"AMBER\",
                    \"test_name\": \"Hematocrit (PCV)\",
                    \"unit\": \"%\",
                    \"value\": \"36.9\"
                },
                {
                    \"reference_range\": \"76 - 100 fL\",
                    \"status\": \"GREEN\",
                    \"test_name\": \"MCV\",
                    \"unit\": \"fL\",
                    \"value\": \"78.1\"
                },
                {
                    \"reference_range\": \"27 - 32 pg\",
                    \"status\": \"RED\",
                    \"test_name\": \"MCH\",
                    \"unit\": \"pg\",
                    \"value\": \"32.6\"
                },
                {
                    \"reference_range\": \"31.5 - 36.5 g/dL\",
                    \"status\": \"GREEN\",
                    \"test_name\": \"MCHC\",
                    \"unit\": \"g/dL\",
                    \"value\": \"32.8\"
                },
                {
                    \"reference_range\": \"11.6 - 15\",
                    \"status\": \"GREEN\",
                    \"test_name\": \"RDW\",
                    \"unit\": \"%\",
                    \"value\": \"13.9\"
                },
                {
                    \"reference_range\": \"7.2 - 12.9 fL\",
                    \"status\": \"AMBER\",
                    \"test_name\": \"MPV\",
                    \"unit\": \"fL\",
                    \"value\": \"11.2\"
                },
                {
                    \"reference_range\": \"4000 - 11000 /cu.mm\",
                    \"status\": \"GREEN\",
                    \"test_name\": \"WBC Count\",
                    \"unit\": \"/cu.mm\",
                    \"value\": \"4567\"
                },
                {
                    \"reference_range\": \"150000 - 450000 /cu.mm\",
                    \"status\": \"GREEN\",
                    \"test_name\": \"Platelets\",
                    \"unit\": \"/cu.mm\",
                    \"value\": \"273000\"
                }
            ],
            \"test_type\": \"HEMOGRAM\"
        },
        {
            \"lab_tests\": [
                {
                    \"reference_range\": \"40 - 70 %\",
                    \"status\": \"GREEN\",
                    \"test_name\": \"Neutrophils\",
                    \"unit\": \"%\",
                    \"value\": \"56.1\"
                },
                {
                    \"reference_range\": \"20 - 45 %\",
                    \"status\": \"GREEN\",
                    \"test_name\": \"Lymphocytes\",
                    \"unit\": \"%\",
                    \"value\": \"20.2\"
                },
                {
                    \"reference_range\": \"2 - 10 %\",
                    \"status\": \"RED\",
                    \"test_name\": \"Monocytes\",
                    \"unit\": \"%\",
                    \"value\": \"14.6\"
                },
                {
                    \"reference_range\": \"0 - 1 %\",
                    \"status\": \"GREEN\",
                    \"test_name\": \"Basophils\",
                    \"unit\": \"%\",
                    \"value\": \"0.1\"
                }
            ],
            \"test_type\": \"DIFFERENTIAL COUNT\"
        },
        {
            \"lab_tests\": [
                {
                    \"reference_range\": \"Normal\",
                    \"status\": \"GREEN\",
                    \"test_name\": \"RBC Morphology\",
                    \"unit\": \"N/A\",
                    \"value\": \"Normocytic Normochromic\"
                },
                {
                    \"reference_range\": \"Normal\",
                    \"status\": \"GREEN\",
                    \"test_name\": \"Platelet Morphology\",
                    \"unit\": \"N/A\",
                    \"value\": \"Adequate\"
                }
            ],
            \"test_type\": \"PERIPHERAL SMEAR EXAMINATION\"
        }
    ]
}"""),
            ],
        ),
        types.Content(
            role="user",
            parts=[
                types.Part.from_text(text=extracted_text),
            ],
        ),
    ]
    generate_content_config = types.GenerateContentConfig(
        temperature=1,
        top_p=0.95,
        top_k=40,
        max_output_tokens=8192,
        response_mime_type="application/json",
        response_schema=genai.types.Schema(
            type = genai.types.Type.OBJECT,
            required = ["metadata", "report"],
            properties = {
                "metadata": genai.types.Schema(
                    type = genai.types.Type.OBJECT,
                    required = ["patient_name", "age", "gender", "lab_name", "report_date"],
                    properties = {
                        "patient_name": genai.types.Schema(
                            type = genai.types.Type.STRING,
                        ),
                        "age": genai.types.Schema(
                            type = genai.types.Type.STRING,
                        ),
                        "gender": genai.types.Schema(
                            type = genai.types.Type.STRING,
                            enum = ["Male", "Female", "Other"],
                        ),
                        "lab_name": genai.types.Schema(
                            type = genai.types.Type.STRING,
                        ),
                        "report_date": genai.types.Schema(
                            type = genai.types.Type.STRING,
                        ),
                    },
                ),
                "report": genai.types.Schema(
                    type = genai.types.Type.ARRAY,
                    items = genai.types.Schema(
                        type = genai.types.Type.OBJECT,
                        required = ["test_type", "lab_tests"],
                        properties = {
                            "test_type": genai.types.Schema(
                                type = genai.types.Type.STRING,
                            ),
                            "lab_tests": genai.types.Schema(
                                type = genai.types.Type.ARRAY,
                                items = genai.types.Schema(
                                    type = genai.types.Type.OBJECT,
                                    required = ["test_name", "value", "unit", "reference_range", "status"],
                                    properties = {
                                        "test_name": genai.types.Schema(
                                            type = genai.types.Type.STRING,
                                        ),
                                        "value": genai.types.Schema(
                                            type = genai.types.Type.STRING,
                                        ),
                                        "unit": genai.types.Schema(
                                            type = genai.types.Type.STRING,
                                        ),
                                        "reference_range": genai.types.Schema(
                                            type = genai.types.Type.STRING,
                                        ),
                                        "status": genai.types.Schema(
                                            type = genai.types.Type.STRING,
                                            enum = ["GREEN", "AMBER", "RED"],
                                        ),
                                    },
                                ),
                            ),
                        },
                    ),
                ),
            },
        ),
        system_instruction=[
            types.Part.from_text(text="""You are an advanced medical data extraction AI designed to process unstructured OCR text from medical lab reports. Your task is to correct errors in test names, values, and reference ranges while ensuring all values align with real-world medical standards. Extract metadata and lab test data in a structured JSON format, strictly following the predefined schema. Assign status labels (GREEN, AMBER, RED) based on whether test values fall within, near, or outside the reference range. Do not add extra fields or modify reference ranges unless corrections are needed for accuracy. Ensure consistent formatting, valid numerical values, and a properly structured JSON output without any deviations."""),
        ],
    )
    try:
        response = client.models.generate_content(
            model=model, contents=contents, config=generate_content_config
        )

        json_response = response.text  # Ensure response is JSON formatted
        parsed_json = json.loads(json_response)  # Convert JSON string to Python dictionary
        return parsed_json

    except json.JSONDecodeError:
        print("Error: Invalid JSON response from the model.")
        return None