aashish1904 commited on
Commit
a0ae69e
1 Parent(s): 7c30dd9

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +365 -0
README.md ADDED
@@ -0,0 +1,365 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ ---
3
+
4
+ language:
5
+ - th
6
+ - en
7
+ base_model: meta-llama/Meta-Llama-3.1-8B
8
+ tags:
9
+ - medical
10
+ - text-generation-inference
11
+ - llama-3.1
12
+ - finetuning
13
+ license: llama3.1
14
+ library_name: transformers
15
+
16
+ ---
17
+
18
+ [![QuantFactory Banner](https://lh7-rt.googleusercontent.com/docsz/AD_4nXeiuCm7c8lEwEJuRey9kiVZsRn2W-b4pWlu3-X534V3YmVuVc2ZL-NXg2RkzSOOS2JXGHutDuyyNAUtdJI65jGTo8jT9Y99tMi4H4MqL44Uc5QKG77B0d6-JfIkZHFaUA71-RtjyYZWVIhqsNZcx8-OMaA?key=xt3VSDoCbmTY7o-cwwOFwQ)](https://hf.co/QuantFactory)
19
+
20
+
21
+ # QuantFactory/Llama-3.1-EIRAI-8B-GGUF
22
+ This is quantized version of [EIRTHAIMED/Llama-3.1-EIRAI-8B](https://huggingface.co/EIRTHAIMED/Llama-3.1-EIRAI-8B) created using llama.cpp
23
+
24
+ # Original Model Card
25
+
26
+ <p align="center">
27
+ <img src="https://cdn-uploads.huggingface.co/production/uploads/66bf1cd096583c59b024a3c5/oG16EyLMfyiqvXrbNPGZd.png" alt="Logo_Website" width="400"/>
28
+ </p>
29
+
30
+ # **Thai Medical Large Language Model**
31
+ **Github** : [Github Evaluate](https://github.com/EIRAI-Thaimedical/EIRAI)<br>
32
+ **PaPer** : <br>
33
+
34
+ ## **Llama-3.1-EIRAI-8B-instruct**
35
+
36
+ **Llama-3.1-EIRAI-8B-instruct**: developed an **8-billion parameter model** specifically tailored for **Thai medical applications**, with expertise in both **Thai medical language** and **English medical terminology**. The model has demonstrated its capabilities through key benchmarks such as **MMLU**, **MedQA**, **PubMedQA**, and **MedMCQA**, as well as Thai language assessments like **ThaiExam**, **M3Exam**, **XNLI**, and **XCOPA**. Additionally, we have created a **Clinically Adapted Model Enhanced test** using the **Thai language** to support **clinical use in hospitals** and to further improve the performance of **Thai medical Retrieval-Augmented Generation (RAG)**.
37
+
38
+ ## Notice
39
+ While **Eir AI Thai Medical LLM** is designed to encode high-quality medical knowledge, it is **not yet optimized for safe, practical use** in real-world medical settings. The model is still in the research phase and should **not be used for clinical decision-making** without further validation, including randomized controlled trials. It is available for researchers to explore the potential of LLMs in medical contexts, but **real-world deployment is not recommended** in its current version.
40
+
41
+ ## Safety and Future Work
42
+ The current version of **Eir AI Thai Medical LLM** is under active development. We advise against using it for medical applications until further testing is completed. Our goal is to continue enhancing the model through **rigorous testing** and **real-world evaluation**, ensuring that it can be safely integrated into healthcare systems in the future.
43
+
44
+ ## Model Overview
45
+ - **Model Architecture:** Meta-Llama-3.1-8B-Instruct
46
+ - **Version:** 1.0
47
+ - **License(s):** [llama3.1](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B/blob/main/LICENSE)
48
+
49
+
50
+ ### Evaluations
51
+
52
+ | Medical Model | Clinical KG | Medical Genetics | Anatomy | Pro Medicine | College Biology | College Medicine | MedQA | PubMedQA | MedMCQA | Avg. |
53
+ |--------------------------|---------------------|---------------------|--------------------|--------------------|--------------------|--------------------|-------------------|-------------------|-------------------|-------------------|
54
+ | **GPT-3.5 Turbo 1106** | 74.7 | 60.2 | 65.9 | 72.0 | 64.73 | 64.73 | 57.71 | 72.66 | 66.0 | 66.6 |
55
+ |Thai LLMs | | | | | | | | | | |
56
+ | **Eir AI-8B** | 75.1 | 80.0 | 69.6 | 76.8 | 77.1 | 66.5 | 64.5 | **79.0** | 58.6 | 71.9 |
57
+ | **Eir AI-8B + Prob** | **83.8** | **89.0** | **83.0** | **84.9** | **89.6** | **75.7** | **69.6** | 78.8 | **67.1** | **80.2** |
58
+ | **Typhoon-v1.5x-8B** | 75.9 | 79.0 | 63.7 | 70.6 | 77.1 | 63.6 | 59.7 | 74.4 | 58.0 | 69.1 |
59
+ | **OpenThaiGPT-beta-7B** | 37.4 | 38.0 | 4.5 | 32.7 | 36.1 | 32.4 | 32.4 | 62.0 | 31.8 | 34.1 |
60
+
61
+ ## Translation Performance Metrics
62
+
63
+ | **Model** | **BLEU Score** | **N-gram Precisions (%)** | **BP** | **Ratio** |
64
+ |-------------------------------|----------------|---------------------------------|---------|-----------|
65
+ | Typhoon-v1.5x-8B-Instruct | 34.42 | 71.3/50.6/38.6/29.6 | 0.764 | 0.788 |
66
+ | Meta Llama 3.1-8B Instruct | 35.74 | 62.8/42.3/31.7/24.1 | 0.946 | 0.948 |
67
+ | **Eir AI-8B** | **61.10** | **76.1/64.6/56.6/50.1** | **1.000**| **1.006** |
68
+ | Eir AI-8B-prob | 47.91 | 74.0/58.0/48.2/40.6 | 0.890 | 0.896 |
69
+
70
+ ## Clinically Adapted Thai Medical Task Performance
71
+
72
+ | Task | GPT-3.5 | Typhoon-v1.5x-8B-instruct | GPT-4o | Eir AI-8B |
73
+ |----------------------------------------|---------|----------------------------|--------|-----------|
74
+ | Named Entity Recognition | 3.26 | 5.55 | 6.34 | **7.08** |
75
+ | Temporal Information Extraction | 3.83 | 5.46 | 6.15 | **7.05** |
76
+ | Paraphrasing | 2.36 | 4.68 | 6.35 | **7.06** |
77
+ | Natural Language Generation | 2.63 | 4.87 | 6.91 | **7.66** |
78
+ | Keyword Extraction | 2.60 | 5.15 | 7.01 | **7.35** |
79
+ | Text Classification | 2.92 | 6.21 | 5.36 | **6.75** |
80
+ | Relation Extraction | 3.29 | 5.94 | 4.37 | **6.92** |
81
+ | Question Answering | 3.70 | 4.92 | 6.11 | **6.82** |
82
+ | Text Summarization | 2.98 | 5.44 | **7.51**| **7.51** |
83
+ | Abbreviation Expansion | 3.99 | 5.96 | 6.24 | **7.82** |
84
+ | Clinical Concept Normalization | 2.67 | 5.63 | 5.82 | **6.55** |
85
+ | Open-ended Question | 3.32 | 5.55 | 6.77 | **7.27** |
86
+ | Multiple-Choice Question | 3.90 | 5.00 | 5.40 | **6.40** |
87
+ | Coreference Resolution | 3.48 | 4.55 | 4.88 | **6.43** |
88
+ | Yes/No Question | 2.71 | 5.86 | 4.86 | **7.38** |
89
+ | Medical Translation | 3.00 | 4.00 | **7.79**| 7.65 |
90
+ | Medical Thai Extraction | 2.81 | 7.16 | **8.62**| 8.16 |
91
+ | Medical ICD Prediction | 2.08 | 3.16 | **8.12**| 6.41 |
92
+ | **Average Score** | 3.05 | 5.33 | 6.38 | **7.11** |
93
+
94
+
95
+ # Prompt Template
96
+
97
+ This model uses `ChatML` prompt template:
98
+
99
+ ```
100
+ <|begin_of_text|><|start_header_id|>system<|end_header_id|>
101
+
102
+ {system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>
103
+
104
+ {prompt}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
105
+
106
+ ````
107
+
108
+ # Example Clinical Adapted ICD 10 Prediction
109
+ ````
110
+ <|begin_of_text|><|start_header_id|>system<|end_header_id|>
111
+
112
+ You are responsible for accurately assigning ICD-10 codes and to diagnose and document medical records.
113
+ Your expertise ensures that healthcare providers are properly reimbursed and that patient care is well-documented.
114
+ In this scenario, you will be presented with a series of medical records and your task is to provide the correct ICD-10 code(s)
115
+ and ICD-9 CM in procedures based on the information provided.
116
+
117
+ <|eot_id|>
118
+ <|start_header_id|>user<|end_header_id|>
119
+
120
+ "Chief Complaint :5วันก่อนมารพ.มีไข้ ไอ มีเสมหะ มีน้ำมูก เหนื่อย ปวดเมื่อยตามตัว \r\n
121
+ Present illness : 5วันก่อนมารพ.มีไข้ ไอ มีเสมหะ มีน้ำมูก เหนื่อย ปวดเมื่อยตามตัว มีน้ำมูก เลือดกำเดาจาากข้างขวา
122
+ ปฏิการกระทบกระแทก ไม่มีเจ็บคอ ไม่มีอาการอ่อนเพลีย มีอาการอ่อนเพลีย ไอมาก ไอตลอด มีอาการระคายคอ ปัสสาวะปกติ ไม่มีถ่ายเหลว
123
+ \r\n\r\nAllergy : |\r\n\r\nOther : no underlying disease\r\n\r\nPlan Treatment Day 1 of hospitalization : admit ward
124
+ \r\n\r\nReview of System { \r\n\r\n General :a thai adult female ,look sickness fatigue dry lip moderate dehydration
125
+ \r\n Skin :no MP rash \r\n Eyes :not pale ,no icteric sclera \r\n Chest :secretion sound in both lung ,no crepitation , no wheezing \r
126
+ \n }
127
+ VitalSign First : {\n
128
+ BP : 117.0/63.0 mmHg\n
129
+ Pulse : 62.0 BPm\n
130
+ Temperature : 37.0 Celsius\n
131
+ Respiratory rate : 20.0\n
132
+ Weight : 50.000 kgs.\n
133
+ Height : 165.0 cm.\n
134
+ Painscore: N/A\n
135
+ O2SAT : 100\n}\n
136
+
137
+ Lab Results: \n
138
+
139
+ Electrolyte:Sodium (Na), Result : 143 mmol/L\r\n
140
+ Electrolyte:Potassium (K),Result : 3.8 mmol/L\r\n
141
+ Electrolyte:Chloride (Cl), Result : 108 mmol/L\r\n
142
+ Electrolyte:Bicarbonate (CO2),Result : 27.0 mmol/L\r\n
143
+ Creatinine (Serum):Creatinine, Result : 0.69 mg/dL\r\n
144
+ Creatinine (Serum):eGFR,Result : 100.41 ml/min/1.73 m^2\r\n
145
+ AST/SGOT:AST/SGOT, Result : 48 U/L\r\n
146
+ ALT/SGPT:ALT/SGPT, Result : 42 U/L\r\n
147
+ CBC:WBC Count,Result : 3.2 10^3/uL\r\n
148
+ CBC:RBC Count, Result : 3.57 10^6/uL\r\n
149
+ CBC:Hemoglobin (Hb), Result : 10.7 g/dL\r\n
150
+ CBC:Hematocrit (HCT),Result : 32.4 %\r\n
151
+ CBC:MCV, Result : 91 fL\r\n
152
+ CBC:MCH, Result : 30.0 pg\r\n
153
+ CBC:MCHC, Result : 33.0 g/dL\r\n
154
+ CBC:RDW-CV,Result : 12.9 %\r\n
155
+ CBC:Platelet Count, Result : 226 10^3/uL\r\n
156
+ CBC:Platelet Estimates, Result : Adequate \r\n
157
+ CBC:Neutrophils,Result : 41 %\r\n
158
+ CBC:Lymphocytes, Result : 46 %\r\n
159
+ CBC:Monocytes, Result : 9 %\r\n
160
+ CBC:Eosinophils, Result : 4 %\r\n
161
+ CBC:Basophils,Result : 0 %\r\n
162
+ COVID-19 Antigen (Rapid Test) For Admit:COVID-19 Antigen (Rapid Test), Result : Positive \r\n
163
+ Influenza A/B screening:Flu A,Result : Negative \r\n
164
+ Influenza A/B screening:Flu B, Result : Negative \n
165
+
166
+ X-ray Results: 15-05-2567 12:56 CHEST (PA, UPRIGHT)\r\n
167
+ \r\nHISTORY: Fever with productive cough and dyspnea for 5 days. \r\nCOMPARISON: None.\r\n
168
+ \r\n
169
+ FINDINGS:\r\n
170
+ TUBES/LINES: None.\r\nLUNGS: Normal lung volumes. Band-like opacity at right apical lung. Clear left lung.\r\n
171
+ PLEURA: No pneumothorax or effusion. Mild bilateral apical pleural thickening. \r\n
172
+ HEART AND MEDIASTINUM: Normal cardiac silhouette. Normal mediastinal contour.\r\n
173
+ BONY STRUCTURES: Mild scoliosis of the thoracic spine. \r\n\r\n
174
+ IMPRESSION:\r\n
175
+ - Band-like opacity at right apical lung, probably plate atelectasis or fibrosis\r\n- Mild bilateral apical pleural thickening"
176
+
177
+ <|eot_id|>
178
+ <|start_header_id|>assistant<|end_header_id|>
179
+
180
+ Let's analyze the provided information and document the appropriate ICD-10 codes based on the patient's symptoms and diagnosis.
181
+
182
+ ### Step-by-Step Analysis:
183
+
184
+ 1. **Understand the Symptoms and Diagnosis**:
185
+ - The patient presents with a range of symptoms including fever, cough with sputum, fatigue, nasal discharge (including blood),
186
+ and dyspnea. The lab tests returned a positive COVID-19 antigen test, while the influenza test was negative.
187
+
188
+ 2. **Identify Key Terms**:
189
+ - Primary key terms include 'COVID-19 positive', 'fever', 'productive cough', 'dyspnea', and pulmonary findings from the X-ray.
190
+
191
+ 3. **Locate the Main Category**:
192
+ - The main categories for the conditions noted are:
193
+ - COVID-19: U07.1
194
+ - Possible acute bronchitis: J20 (based on the productive cough)
195
+ - Other symptoms such as dizziness: R42
196
+ - Evidence of acute respiratory infection: J06.9
197
+
198
+ 4. **Identify Specific Codes**:
199
+ - Based on the diagnosis:
200
+ - COVID-19 confirmed case: U07.1
201
+ - Acute bronchitis: J20.9 (considering unspecified acute bronchitis)
202
+ - Dizziness (if noted as significant and problematic): R42
203
+ - Unspecified acute upper respiratory infection: J06.9
204
+
205
+ 5. **Check for Additional Codes or Modifiers**:
206
+ - Considering the details provided, the primary diagnosis is the confirmed COVID-19 infection, while bronchiolitis
207
+ and dizziness might be secondary diagnoses.
208
+
209
+ 6. **Validate the Code Selection**:
210
+ - Confirm that the selected codes reflect the patient's symptoms accurately, including acute bronchitis
211
+ and any other significant symptoms presented.
212
+
213
+ 7. **Document the ICD-10 Code**:
214
+ - The ICD-10 codes are documented for all noted conditions accordingly."
215
+
216
+ ```json
217
+ Answer: {
218
+ "Primary_Diagnosis": {
219
+ "Code": "U071",
220
+ "Description": "COVID-19, virus identified"
221
+ },
222
+ "Secondary_Diagnosis": [
223
+ {
224
+ "Code": "J20",
225
+ "Description": "Acute bronchitis due to unspecified organism"
226
+ },
227
+ {
228
+ "Code": "R42",
229
+ "Description": "Dizziness and giddiness"
230
+ },
231
+ {
232
+ "Code": "J069",
233
+ "Description": "Acute upper respiratory infection, unspecified"
234
+ }
235
+ ],
236
+ "ICD9_Procedure": []
237
+ }
238
+
239
+
240
+ ````
241
+
242
+
243
+
244
+ # Example Clinical Adapted Thai Medical Extraction
245
+ ````
246
+ <|begin_of_text|><|start_header_id|>system<|end_header_id|>
247
+
248
+ Task : Extract input the following patient information into output format Tone: the following medical text into
249
+ Thai in a fluent and elegant style.
250
+ Output Format.1.Age: \n2.Gender: \n3.Weight :\n4.Height : \n5.Chief Complaint: \n6.Symptoms and Signs: \n7.Medical History: \n
251
+ 8.Current Medications: \n9.Laboratory Results: \n10.Imaging Findings: \n11.Allergy: \n12.Drug Allergy:
252
+
253
+
254
+ <|eot_id|>
255
+ <|start_header_id|>user<|end_header_id|>
256
+ ผู้ป่วยของเราเป็นชายถนัดทั้งสองมือ อายุ 43 ปี มีประวัติการชักที่ไม่สามารถควบคุมได้มาเป็นเวลา 20 ปี ลักษณะการชักของเขามักจะรวมถึงการรู้สึกร้อนวูบวาบและอาการทางประสาทสัมผัสอื่น ๆ
257
+ ที่พัฒนาไปสู่การเคลื่อนไหวของกล้ามเนื้อที่มีจุดศูนย์กลางส่วนใหญ่ทางด้านขวา การตรวจหาสาเหตุขอ��การชักรวมถึงการถ่ายภาพด้วยคลื่นแม่เหล็กไฟฟ้า (MRI) ซึ่งเผยให้เห็นเนื้องอกไขมันขนาดใหญ่ที่เส้นกลางสมอง
258
+ การพัฒนาไม่สมบูรณ์ของคอร์ปัสคาโลซัมบางส่วน และรอยโรคที่อยู่ใกล้เคียงในสมองส่วนหน้าซ้ายที่คาดว่าจะเป็นเนื้องอกกลีอาล (glial neoplasm) ตามลักษณะภาพถ่ายทางรังสี
259
+ รอยโรคในสมองส่วนหน้าซ้ายด้านหน้าและตรงกลางประกอบด้วยการกลายเป็นหินปูนแบบเป็นก้อนพร้อมการเพิ่มขึ้นของสัญญาณ FLAIR ที่กว้างขวางซึ่งเกี่ยวข้องกับไจรัสซิงกูเลตทั้งสองข้างและสมองส่วนหน้าซ้าย
260
+ (รูปที่ ).\n\nการจัดการทางการแพทย์ล้มเหลวในการควบคุมการชักของเขาและเขาถูกส่งต่อเพื่อหาทางเลือกในการรักษาด้วยการผ่าตัด รอยโรคที่เพิ่มขึ้นถูกสังเกตด้วยการถ่ายภาพเพิ่มเติมและขอบเขตของอาการบวมน้ำก็เพิ่มขึ้นด้วย
261
+ ความกังวลเกี่ยวกับการพัฒนาเนื้องอกกลีอาลที่เพิ่มขึ้นและการควบคุมการชักที่ไม่ดีทำให้มีการแนะนำให้ทำการผ่าตัด
262
+ การตัดสินใจถูกทำขึ้นเพื่อดำเนินการผ่าตัดนำทางด้วยระบบประสาทเพื่อตัดมวลที่เพิ่มขึ้นในสมองส่วนหน้าซ้ายและการตัดสมองส่วนหน้าบางส่วนโดยใช้การตรวจคลื่นไฟฟ้าสมองระหว่างการผ่าตัด
263
+ (intraoperative electroencephalogram - EEG), การทำแผนที่คอร์ติคอล (cortical mapping) และการตรวจวัดศักย์ไฟฟ้าที่เกิดจากการกระตุ้นประสาทรับความรู้สึก
264
+ (somatosensory evoked potentials - SSEP)\n\nตัวอย่างที่ส่งไปตรวจทางพยาธิวิทยาแบบแช่แข็งในระหว่างการผ่าตัดพบว่ามีเส้นใยโรเซนธาล (Rosenthal fibers)
265
+ และการกลายเป็นหินปูนแบบเป็นจุดซึ่งคาดว่าจะเป็นเนื้องอกกลีอาล การประเมินทางพยาธิวิทยาแบบถาวรเผยให้เห็นเนื้องอกไขมัน (lipoma) และความผิดปกติของคอร์ติคอลแบบเฉพาะจุด
266
+ (focal cortical dysplasia) แบบ Palmini Type IA ในสมองที่อยู่ใกล้เคียง ความผิดปกติเล็กน้อยของโครงสร้างคอร์ติคอลและการเกิดกลีโอซิส (gliosis)
267
+ ในเนื้อสมองขาวที่เกี่ยวข้องสามารถเห็นได้ในคราบสีฮีมาโทซิลินและอีโอซิน (hematoxylin and eosin - H&E) และคราบสีโปรตีนกรดกลีอาลไฟบริลลารี (glial fibrillary acidic protein - GFAP)
268
+ และการย้อมสีโปรตีนเส้นประสาท (neurofilament protein - NFP) ในบริเวณที่เกิดกลีโอซิสไม่แสดงหลักฐานของเซลล์ประสาทที่ผิดรูป เซลล์ประสาทขนาดใหญ่ หรือเซลล์ลูกโป่ง (รูปที่ ).\n\n
269
+ การถ่ายภาพหลังการผ่าตัด (รูปที่ ) แสดงให้เห็นการตัดรอยโรคที่เพิ่มขึ้นใ��ล้เคียงทั้งหมดในสมองส่วนหน้าซ้ายและไจรัสซิงกูเลต การลดลงอย่างมากของอาการบวมน้ำที่เกิดจากหลอดเลือดถูกสังเกตในภาพ
270
+ FLAIR เนื้องอกไขมันที่เส้นกลางสมองไม่เปลี่ยนแปลง\n\nการฟื้นตัวหลังการตัดรวมถึงอาการเสริมในบริเวณมอเตอร์ซินโดรมระดับปานกลางรวมถึงอาการอ่อนแรงครึ่งซีก การพูดล่าช้า และการเปลี่ยนแปลงทางอารมณ์
271
+ การเปลี่ยนแปลงเหล่านี้ค่อยๆ หายไปและพบว่าผู้ป่วยมีความสมบูรณ์ทางระบบประสาทในการติดตามผลที่คลินิก การควบคุมการชักเกิดขึ้นทันทีหลังการตัด ยาต้านการชักถูกใช้อย่างต่อเนื่องและค่อยๆ
272
+ ถูกถอดออกในเดือนต่อมาโดยไม่มีการกลับมาเป็นซ้ำ ผู้ป่วยไม่มีการชักเป็นเวลา 3 ปีหลังการตัด การถ่ายภาพติดตามผล 18 เดือนพบการเปลี่ยนแปลง T2/FLAIR ที่คงที่พร้อมกับบริเวณเล็ก ๆ
273
+
274
+
275
+ <|eot_id|>
276
+ <|start_header_id|>assistant<|end_header_id|>
277
+ 1. อายุ: 43 ปี\n
278
+ 2. เพศ: ชาย\n
279
+ 3. น้ำหนัก: ไม่ระบุ\n
280
+ 4. ส่วนสูง: ไม่ระบุ\n
281
+ 5. อาการหลัก: การชักที่ไม่สามารถควบคุมได้\n
282
+ 6. อาการและสัญญาณ: รู้สึกร้อนวูบวาบ, อาการทางประสาทสัมผัส, การเคลื่อนไหวของกล้ามเนื้อที่มีจุดศูนย์กลางส่วนใหญ่ทางด้านขวา\n
283
+ 7. ประวัติการรักษา: การจัดการทางการแพทย์ล้มเหลวในการควบคุมการชัก\n
284
+ 8. ยาที่ใช้ในปัจจุบัน: ยาต้านการชัก\n9. ผลการตรวจทางห้องปฏิบัติการ: ไม่ระบุ\n
285
+ 10. ผลการตรวจภาพ: MRI แสดงเนื้องอกไขมันขนาดใหญ่ที่เส้นกลางสมอง, การพัฒนาไม่สมบูรณ์ของคอร์ปัสคาโลซัมบางส่วน, รอยโรคที่อยู่ใกล้เคียงในสมองส่วนหน้าซ้ายที่คาดว่าจะเป็นเนื้องอกกลีอาล\n
286
+ 11. อาการแพ้: ไม่ระบุ\n
287
+ 12. อาการแพ้ยา: ไม่ระบุ\n
288
+ การผ่าตัดนำทางด้วยระบบประสาทเพื่อตัดมวลที่เพิ่มขึ้นในสมองส่วนหน้าซ้ายและการตัดสมองส่วนหน้าบางส่วนได้ถูกดำเนินการ
289
+ และผลการตรวจทางพยาธิวิทยาเผยให้เห็นว่าเป็นเนื้องอกไขมันและความผิดปกติของคอร์ติคอลแบบเฉพาะจุด (focal cortical dysplasia) แบบ Palmini Type IA
290
+ ในสมองที่อยู่ใกล้เคียง การฟื้นตัวหลังการผ่าตัดรวมถึงอาการเสริมในบริเวณมอเตอร์ซินโดรมระดับปานกลาง แต่การควบคุมการชักเกิดขึ้นทันทีหลังการผ่าตัดและผู้ป่วยไม่มีการชักเป็นเวลา
291
+ 3 ปีหลังการผ่าตัด การถ่ายภาพติดตามผล 18 เดือนพบการเปลี่ยนแปลง T2/FLAIR ที่คงที่พร้อมกับบริเวณเล็ก ๆ ที่เพิ่มขึ้นของการเพิ่มขึ้นใกล้เคียงท���่ไม่เปลี่ยนแปลง.
292
+ ````
293
+
294
+
295
+
296
+
297
+ # How to use
298
+
299
+
300
+ ```python
301
+
302
+ from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
303
+ import torch
304
+
305
+ model_id = "EIRTHAIMED/Llama-3.1-EIRAI-8B"
306
+
307
+ nf4_config = BitsAndBytesConfig(
308
+ load_in_4bit=True,
309
+ bnb_4bit_quant_type="nf4",
310
+ bnb_4bit_use_double_quant=True,
311
+ bnb_4bit_compute_dtype=torch.bfloat16
312
+ )
313
+
314
+ # Load the base model
315
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
316
+ model = AutoModelForCausalLM.from_pretrained(
317
+ model_id,
318
+ torch_dtype=torch.bfloat16,
319
+ # quantization_config=nf4_config, # uncomment this line for 4 bit loading
320
+ device_map="auto",
321
+ attn_implementation="flash_attention_2"
322
+ )
323
+
324
+ messages = [
325
+ {"role": "system", "content": "You are an expert medical assistant named EIR , developed by EIR Thai Medical LLM. You are to be a helpful, respectful, and honest assistant."},
326
+ {"role": "user", "content": "การใช้ clinical tracer มีบทบาทอย่างไรในการพัฒนาคุณภาพการดูแลผู้ป่วย?"}
327
+ ]
328
+
329
+ input = tokenizer.apply_chat_template(
330
+ messages,
331
+ tokenize = True,
332
+ add_generation_prompt = True, # Must add for generation
333
+ return_tensors = "pt",
334
+ ).to("cuda")
335
+
336
+
337
+ from transformers import TextStreamer
338
+ text_streamer = TextStreamer(tokenizer, skip_prompt = True)
339
+ _ = model.generate(input, streamer = text_streamer, max_new_tokens = 1500, do_sample=True, temperature=0.01, top_k=100, top_p=0.95)
340
+
341
+
342
+ ```
343
+
344
+
345
+ ```
346
+ @article{EirAI,
347
+ title={Eir: Thai Medical Large Language Models},
348
+ author={Yutthakorn Thiprak and Rungtam Ngodngamthaweesuk and Songtam Ngodngamtaweesuk, MD},
349
+ year={2024},
350
+ journal={arXiv preprint arXiv:2409.08523},
351
+ url={https://arxiv.org/abs/2409.08523}
352
+ }
353
+ ```
354
+
355
+ ---
356
+ **Thank you very much**
357
+ Asst.Prof.Dr. Ekapol Chuangsuwanich and Praj Bhargava @Meta Research Engineer, for your valuable endorsement of our preprint paper on arXiv.
358
+
359
+ **Thank you**
360
+ Draft Reviewer Report
361
+ [Kullawat Chaowanawatee](https://www.computing.psu.ac.th/profile/index.php?staffid=coc0051) and [Dr. Jakapan Suaboot](https://www.computing.psu.ac.th/profile/index.php?staffid=coc0056) from Prince of Songkla University, Phuket Campus
362
+ <br>
363
+ Draft Industry Reviewer Report
364
+ [Mr. Piyawat Maneenual](https://ieeexplore.ieee.org/author/37086452350) ,Assistant IT Manager ,Thonburi Rajyindee Hospital<br>
365
+