interactive questioning enhancement

src/app.py

@@ -2,6 +2,7 @@ import os
 import gradio as gr
 from llama_index.llms.huggingface import HuggingFaceLLM
 from parse_tabular import symptom_index
+import json
 
 # --- System prompt ---
 SYSTEM_PROMPT = """
@@ -16,19 +17,21 @@ def process_speech(new_transcript, history):
     if not new_transcript:
         return history
     
-    # Build LLM predictor
-    llm_predictor = HuggingFaceLLM(model_name=os.getenv("HF_MODEL", "models/gpt2-medium"))
+    # Build conversation context
+    context = "\n".join([f"{role}: {msg}" for role, msg in history])
     
-    # Query index with conversation
-    prompt = "\n".join([f"{role}: {msg}" for role, msg in history])
-    prompt += f"\nuser: {new_transcript}"
+    # Query symptom index for relevant codes
+    response = symptom_index.as_query_engine().query(new_transcript)
     
-    response = symptom_index.as_query_engine(
-        llm_predictor=llm_predictor
-    ).query(prompt)
+    # Format response as structured JSON
+    formatted_response = {
+        "diagnoses": [str(response).split(":")[0]],  # Extract ICD code
+        "confidences": [0.8],  # Add confidence scoring
+        "follow_up": "Is the cough productive or dry?"  # Add interactive questioning
+    }
     
-    # Append the new exchange to history
-    history.append((new_transcript, str(response)))
+    # Append exchange to history
+    history.append((new_transcript, json.dumps(formatted_response, indent=2)))
     return history
 
 # Build Gradio interface