json

.gitattributes

@@ -33,3 +33,13 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+# Local development files
+*.pyc
+__pycache__/
+.env
+*.log
+.DS_Store
+
+# Never commit API keys!
+*.key
+secrets.txt

app.py

@@ -2,8 +2,6 @@
 Module 1: Cross-Cultural Semantic Translator MVP
 =================================================
 A medical AI platform for translating cultural pain metaphors into structured medical ontologies.
-
-Deployed on Hugging Face Spaces.
 """
 
 import gradio as gr
@@ -12,20 +10,15 @@ import os
 from typing import Dict, Tuple, Optional
 
 # ============================================================================
-# CONFIGURATION - 从环境变量读取
+# CONFIGURATION
 # ============================================================================
 
-# SECURITY: NEVER hardcode API keys in public repos
 OPENAI_API_KEY = os.getenv("OPENAI_API_KEY", "")
-
-# Transcription mode: Force API mode on Hugging Face (no GPU access)
 TRANSCRIPTION_MODE = "api"
-
-# OpenAI model for analysis
-OPENAI_MODEL = "gpt-4.1"  # Use newer model name
+OPENAI_MODEL = "gpt-4-turbo-preview"
 
 # ============================================================================
-# IMPORTS AND SETUP
+# SETUP
 # ============================================================================
 
 try:
@@ -36,24 +29,21 @@ except ImportError:
     client = None
 
 # ============================================================================
-# SYSTEM PROMPT FOR LLM
+# SYSTEM PROMPT
 # ============================================================================
 
 MEDICAL_ANTHROPOLOGIST_PROMPT = """You are an expert Medical Anthropologist. Your goal is to translate cultural pain metaphors into structured medical ontologies. Do NOT act as a doctor making a final diagnosis. Analyze the patient's transcript and output a strict JSON object with these exact keys: 'literal_translation', 'metaphor_mapping', 'mcgill_pain_ontology', 'psychological_and_stoicism_flags', 'physician_action_note'. Make sure to include English and original language in metaphor_mapping for reference."""
 
 # ============================================================================
-# AUDIO TRANSCRIPTION FUNCTION
+# TRANSCRIPTION
 # ============================================================================
 
 def transcribe_audio(audio_path: Optional[str]) -> Tuple[str, str]:
-    """
-    Transcribe audio using OpenAI Whisper API.
-    """
     if audio_path is None:
-        return "", "⚠️ No audio recorded. Please record audio first."
+        return "", "⚠️ No audio recorded."
     
     if client is None:
-        return "", "❌ OpenAI client not initialized. API key missing."
+        return "", "❌ OpenAI client not initialized."
     
     try:
         with open(audio_path, "rb") as audio_file:
@@ -62,32 +52,17 @@ def transcribe_audio(audio_path: Optional[str]) -> Tuple[str, str]:
                 file=audio_file,
                 response_format="text"
             )
-        transcription = transcript.strip()
-        status = f"✓ Transcribed via OpenAI Whisper API"
-        
-        if not transcription:
-            return "", "⚠️ Transcription is empty. Please check your audio quality."
-        
-        return transcription, status
-    
+        return transcript.strip(), "✓ Transcribed via OpenAI Whisper API"
     except Exception as e:
-        error_msg = f"❌ Transcription error: {str(e)}"
-        print(error_msg)
-        return "", error_msg
+        return "", f"❌ Transcription error: {str(e)}"
 
 # ============================================================================
-# LLM ANALYSIS FUNCTION
+# LLM ANALYSIS
 # ============================================================================
 
 def analyze_with_llm(transcription: str) -> Tuple[str, str]:
-    """
-    Send transcription to OpenAI API for medical anthropological analysis.
-    """
-    if not transcription or transcription.strip() == "":
-        return "<div style='padding: 20px; color: #ffc107;'>⚠️ No transcription to analyze.</div>", "{}"
-    
-    if client is None:
-        return "<div style='padding: 20px; color: #ff6b6b;'>❌ OpenAI client not initialized. Please set OPENAI_API_KEY in Space secrets.</div>", "{}"
+    if not transcription or not client:
+        return "<div style='padding: 20px; color: #ff6b6b;'>❌ Cannot analyze</div>", "{}"
     
     try:
         response = client.chat.completions.create(
@@ -101,58 +76,178 @@ def analyze_with_llm(transcription: str) -> Tuple[str, str]:
         )
         
         json_text = response.choices[0].message.content
-        
-        if not json_text or json_text.strip() == "":
-            return "<div style='padding: 20px; color: #ff6b6b;'>❌ Empty response from LLM</div>", "{}"
-        
-        try:
-            parsed_json = json.loads(json_text)
-        except json.JSONDecodeError as je:
-            error_html = f"""
-            <div style='padding: 20px; background-color: #f8d7da; border-left: 5px solid #dc3545; border-radius: 8px;'>
-                <h3 style='color: #721c24;'>⚠️ JSON Parse Error</h3>
-                <p style='color: #721c24;'>{str(je)}</p>
-            </div>
-            """
-            return error_html, json_text
-        
+        parsed_json = json.loads(json_text)
         formatted_output = format_json_for_display(parsed_json)
+        
         return formatted_output, json_text
     
     except Exception as e:
         import traceback
-        error_details = traceback.format_exc()
         error_html = f"""
         <div style='padding: 20px; background-color: #f8d7da; border-left: 5px solid #dc3545; border-radius: 8px;'>
-            <h3 style='color: #721c24;'>❌ LLM Analysis Error</h3>
-            <pre style='color: #721c24; font-size: 12px; overflow-x: auto;'>{error_details}</pre>
+            <h3 style='color: #721c24;'>❌ Error</h3>
+            <pre style='color: #721c24; font-size: 12px; overflow-x: auto;'>{traceback.format_exc()}</pre>
         </div>
         """
         return error_html, "{}"
 
 # ============================================================================
-# JSON FORMATTING FOR DISPLAY
+# JSON FORMATTING - 完整版本从 semantic_translator_mvp.py 复制
 # ============================================================================
 
 def format_json_for_display(data: Dict) -> str:
-    """Format the JSON output into a human-readable medical report."""
-    # [保留你原来的 format_json_for_display 函数的完整代码]
-    # 这里为了简洁省略，实际部署时复制完整函数
+    """Format JSON into human-readable medical report"""
+    
     html_parts = ['''
-    <div style="font-family: 'Segoe UI', Arial, sans-serif; padding: 30px; background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); border-radius: 15px; color: #ffffff; box-shadow: 0 10px 25px rgba(0,0,0,0.2); line-height: 1.8;">
+    <div style="
+        font-family: 'Segoe UI', Arial, sans-serif; 
+        padding: 30px; 
+        background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+        border-radius: 15px;
+        color: #ffffff;
+        box-shadow: 0 10px 25px rgba(0,0,0,0.2);
+        line-height: 1.8;
+    ">
     ''']
     
-    # [完整的格式化逻辑...]
+    # Debug section
+    import json
+    raw_json = json.dumps(data, indent=2, ensure_ascii=False)
+    html_parts.append(f'''
+    <details style="margin-bottom: 20px; padding: 15px; background-color: rgba(0, 0, 0, 0.2); border-radius: 8px;">
+        <summary style="cursor: pointer; font-weight: bold; color: #ffd700;">🔍 Debug: Raw JSON</summary>
+        <pre style="margin-top: 10px; padding: 10px; background-color: rgba(0, 0, 0, 0.3); border-radius: 5px; overflow-x: auto; font-size: 12px; color: #e0e0e0;">{raw_json}</pre>
+    </details>
+    ''')
+    
+    # 1. Literal Translation
+    if 'literal_translation' in data:
+        html_parts.append(f'''
+        <div style="margin-bottom: 25px; padding: 20px; background-color: rgba(255,255,255,0.15); border-left: 5px solid #ffd700; border-radius: 10px;">
+            <h2 style="margin: 0 0 15px 0; color: #ffd700; font-size: 22px; font-weight: 700;">📝 Patient's Description</h2>
+            <p style="margin: 0; font-size: 16px; color: #ffffff; font-style: italic;">"{data['literal_translation']}"</p>
+        </div>
+        ''')
+    
+    # 2. Metaphor Mapping
+    if 'metaphor_mapping' in data:
+        metaphor = data['metaphor_mapping']
+        html_parts.append('''
+        <div style="margin-bottom: 25px; padding: 20px; background-color: rgba(255,255,255,0.15); border-left: 5px solid #4fc3f7; border-radius: 10px;">
+            <h2 style="margin: 0 0 15px 0; color: #4fc3f7; font-size: 22px; font-weight: 700;">🔗 Cultural Context</h2>
+        ''')
+        
+        def render_value(val, indent=0):
+            margin_left = indent * 20
+            if isinstance(val, dict):
+                items = []
+                for k, v in val.items():
+                    k_readable = k.replace('_', ' ').title()
+                    items.append(f'<div style="margin: 8px 0 8px {margin_left}px;"><strong style="color: #81d4fa;">{k_readable}:</strong>{render_value(v, indent+1)}</div>')
+                return ''.join(items)
+            elif isinstance(val, list):
+                if not val:
+                    return '<span style="margin-left: 10px; color: #e0e0e0;">None</span>'
+                items_html = '<ul style="margin: 5px 0; padding-left: 20px; color: #e0e0e0;">'
+                for item in val:
+                    items_html += f'<li style="margin: 5px 0;">{render_value(item, indent) if isinstance(item, (dict, list)) else str(item)}</li>'
+                items_html += '</ul>'
+                return items_html
+            else:
+                return f'<span style="margin-left: 10px; font-size: 15px; color: #ffffff;">{str(val)}</span>'
+        
+        html_parts.append(render_value(metaphor))
+        html_parts.append('</div>')
+    
+    # 3. McGill Pain Ontology
+    if 'mcgill_pain_ontology' in data:
+        mcgill = data['mcgill_pain_ontology']
+        html_parts.append('''
+        <div style="margin-bottom: 25px; padding: 20px; background-color: rgba(255,255,255,0.15); border-left: 5px solid #ff6b6b; border-radius: 10px;">
+            <h2 style="margin: 0 0 15px 0; color: #ff6b6b; font-size: 22px; font-weight: 700;">🏥 McGill Pain Assessment</h2>
+        ''')
+        
+        field_icons = {
+            'location': '📍',
+            'temporal_pattern': '⏱️',
+            'intensity': '📊',
+            'quality_descriptors': '💭',
+            'associated_symptoms_to_query': '🔍',
+            'functional_impact_to_query': '🚶',
+            'pain_or_sensory_type': '🩺'
+        }
+        
+        def render_mcgill(val, indent=1):
+            margin_left = indent * 20
+            if isinstance(val, dict):
+                items = []
+                for k, v in val.items():
+                    k_readable = k.replace('_', ' ').title()
+                    items.append(f'<div style="margin: 5px 0 5px {margin_left}px;"><em style="color: #ffd4d4;">{k_readable}:</em>{render_mcgill(v, indent+1)}</div>')
+                return ''.join(items)
+            elif isinstance(val, list):
+                if not val:
+                    return '<span style="margin-left: 10px; color: #e0e0e0;">None specified</span>'
+                return '<span style="margin-left: 10px; color: #ffffff;">' + ', '.join(str(v) for v in val) + '</span>'
+            else:
+                return f'<span style="margin-left: 10px; color: #ffffff;">{str(val)}</span>'
+        
+        if isinstance(mcgill, list):
+            for item in mcgill:
+                if isinstance(item, dict):
+                    for key, value in item.items():
+                        key_readable = key.replace('_', ' ').title()
+                        icon = field_icons.get(key, '•')
+                        html_parts.append(f'<div style="margin-bottom: 15px; padding: 12px; background-color: rgba(255,255,255,0.1); border-radius: 8px;"><strong style="color: #ffcccb; font-size: 16px;">{icon} {key_readable}:</strong>{render_mcgill(value)}</div>')
+                else:
+                    html_parts.append(f'<div style="margin-bottom: 15px; padding: 12px; background-color: rgba(255,255,255,0.1); border-radius: 8px;"><p style="margin: 0; font-size: 15px; color: #ffffff;">{str(item)}</p></div>')
+        elif isinstance(mcgill, dict):
+            for key, value in mcgill.items():
+                key_readable = key.replace('_', ' ').title()
+                icon = field_icons.get(key, '•')
+                html_parts.append(f'<div style="margin-bottom: 15px; padding: 12px; background-color: rgba(255,255,255,0.1); border-radius: 8px;"><strong style="color: #ffcccb; font-size: 16px;">{icon} {key_readable}:</strong>{render_mcgill(value)}</div>')
+        else:
+            html_parts.append(f'<div style="margin-bottom: 15px; padding: 12px; background-color: rgba(255,255,255,0.1); border-radius: 8px;"><p style="margin: 0; font-size: 15px; color: #ffffff;">{str(mcgill)}</p></div>')
+        
+        html_parts.append('</div>')
+    
+    # 4. Psychological Flags
+    if 'psychological_and_stoicism_flags' in data:
+        psych = data['psychological_and_stoicism_flags']
+        html_parts.append('''
+        <div style="margin-bottom: 25px; padding: 20px; background-color: rgba(255,255,255,0.15); border-left: 5px solid #9c27b0; border-radius: 10px;">
+            <h2 style="margin: 0 0 15px 0; color: #ce93d8; font-size: 22px; font-weight: 700;">🧠 Psychological Assessment</h2>
+        ''')
+        
+        for key, value in psych.items():
+            key_readable = key.replace('_', ' ').title()
+            if isinstance(value, dict):
+                html_parts.append(f'<p style="margin: 10px 0; font-size: 15px;"><strong style="color: #ce93d8;">{key_readable}:</strong></p>')
+                for sub_key, sub_value in value.items():
+                    sub_key_readable = sub_key.replace('_', ' ').title()
+                    html_parts.append(f'<p style="margin: 5px 0 5px 20px; font-size: 14px; color: #e0e0e0;">• {sub_key_readable}: {sub_value}</p>')
+            else:
+                html_parts.append(f'<p style="margin: 10px 0; font-size: 15px;"><strong style="color: #ce93d8;">{key_readable}:</strong> <span style="color: #ffffff;">{value}</span></p>')
+        
+        html_parts.append('</div>')
+    
+    # 5. Physician Action Note
+    if 'physician_action_note' in data:
+        html_parts.append(f'''
+        <div style="padding: 20px; background-color: rgba(255,255,255,0.2); border: 3px solid #4caf50; border-radius: 10px;">
+            <h2 style="margin: 0 0 15px 0; color: #a5d6a7; font-size: 22px; font-weight: 700;">⚕️ Clinical Recommendations</h2>
+            <p style="margin: 0; font-size: 16px; color: #ffffff; line-height: 1.9;">{data['physician_action_note']}</p>
+        </div>
+        ''')
     
     html_parts.append('</div>')
     return ''.join(html_parts)
 
 # ============================================================================
-# MAIN PROCESSING FUNCTION
+# MAIN PROCESSING
 # ============================================================================
 
 def process_patient_audio(audio) -> Tuple[str, str, str]:
-    """Main processing pipeline"""
     try:
         transcription, trans_status = transcribe_audio(audio)
         
@@ -168,91 +263,66 @@ def process_patient_audio(audio) -> Tuple[str, str, str]:
     
     except Exception as e:
         import traceback
-        error_details = traceback.format_exc()
         error_html = f"""
         <div style='padding: 20px; background-color: #f8d7da; border-left: 5px solid #dc3545; border-radius: 8px;'>
             <h3 style='color: #721c24;'>❌ Unexpected Error</h3>
-            <pre style='color: #721c24; font-size: 12px; overflow-x: auto;'>{error_details}</pre>
+            <pre style='color: #721c24; font-size: 12px; overflow-x: auto;'>{traceback.format_exc()}</pre>
         </div>
         """
         return "❌ Processing error", "Error during processing", error_html
 
 # ============================================================================
-# GRADIO UI SETUP
+# GRADIO UI
 # ============================================================================
 
 def create_ui():
-    """Create the Gradio interface"""
-    
     with gr.Blocks(title="Medical AI Semantic Translator", theme=gr.themes.Soft()) as app:
-        gr.Markdown(
-            """
-            # 🏥 Module 1: Cross-Cultural Semantic Translator
-            ### Translating Cultural Pain Metaphors into Medical Ontologies
-            
-            **Instructions:** Record your audio description of pain symptoms, then click Analyze.
-            
-            ⚠️ **Note:** This demo uses OpenAI's API. The Space owner must configure API keys.
-            """
-        )
+        gr.Markdown("""
+        # 🏥 Module 1: Cross-Cultural Semantic Translator
+        ### Translating Cultural Pain Metaphors into Medical Ontologies
+        
+        **Instructions:** Record your audio description, then click Analyze.
+        """)
         
         status_output = gr.Textbox(label="Status", interactive=False, lines=1)
         
         with gr.Row():
             with gr.Column(scale=1):
                 gr.Markdown("### 🎤 Audio Input")
-                audio_input = gr.Audio(
-                    sources=["microphone"],
-                    type="filepath",
-                    label="Record Your Pain Description"
-                )
+                audio_input = gr.Audio(sources=["microphone"], type="filepath", label="Record Your Pain Description")
                 submit_btn = gr.Button("🔍 Analyze", variant="primary", size="lg")
                 
                 gr.Markdown("### 📄 Transcription")
-                transcription_output = gr.Textbox(
-                    label="Whisper Transcription Output",
-                    interactive=False,
-                    lines=8
-                )
+                transcription_output = gr.Textbox(label="Whisper Transcription", interactive=False, lines=8)
             
             with gr.Column(scale=1):
                 gr.Markdown("### 🤖 AI Medical Anthropologist Analysis")
-                analysis_output = gr.HTML(
-                    label="Structured Medical Ontology",
-                    value='<div style="padding: 20px; text-align: center; color: #6c757d;">Analysis results will appear here...</div>'
-                )
+                analysis_output = gr.HTML(value='<div style="padding: 20px; text-align: center; color: #6c757d;">Analysis results will appear here...</div>')
         
-        gr.Markdown(
-            f"""
-            ---
-            **Configuration:** Transcription: `API` | LLM Model: `{OPENAI_MODEL}`
-            
-            **Deployed on:** [Hugging Face Spaces](https://huggingface.co/spaces/DIrtyCha/Module1demo)
-            """
-        )
+        gr.Markdown(f"""
+        ---
+        **Configuration:** `API` mode | `{OPENAI_MODEL}`
+        **Deployed on:** [Hugging Face Spaces](https://huggingface.co/spaces/DIrtyCha/Module1demo)
+        """)
         
-        submit_btn.click(
-            fn=process_patient_audio,
-            inputs=[audio_input],
-            outputs=[status_output, transcription_output, analysis_output]
-        )
+        submit_btn.click(fn=process_patient_audio, inputs=[audio_input], outputs=[status_output, transcription_output, analysis_output])
     
     return app
 
 # ============================================================================
-# MAIN ENTRY POINT
+# MAIN
 # ============================================================================
 
 if __name__ == "__main__":
-    print("="  * 70)
-    print("🚀 Medical AI Semantic Translator MVP - Hugging Face Spaces")
+    print("=" * 70)
+    print("🚀 Medical AI Semantic Translator MVP")
     print("=" * 70)
     
     if not OPENAI_API_KEY:
-        print("⚠️  WARNING: OPENAI_API_KEY not set in Space secrets!")
-        print("   Go to: Settings → Repository Secrets → Add OPENAI_API_KEY")
+        print("⚠️  WARNING: OPENAI_API_KEY not set!")
+        print("   Go to Settings → Repository Secrets")
     else:
-        print("✅ OpenAI API key loaded from environment")
+        print("✅ OpenAI API key loaded")
     
     print("=" * 70)