Update app.py

app.py

@@ -5,102 +5,126 @@ from langchain_community.vectorstores import FAISS
 from langchain_huggingface import HuggingFaceEmbeddings
 from langchain_text_splitters import RecursiveCharacterTextSplitter
 from langchain_groq import ChatGroq
-from langchain.chains import RetrievalQA
-from langchain.prompts import PromptTemplate
+from langchain_core.prompts import ChatPromptTemplate
+from langchain_core.runnables import RunnablePassthrough
+from langchain_core.output_parsers import StrOutputParser
 
-# --- 1. CONFIGURATION & SECRETS ---
-# Note: On Hugging Face, add GROQ_API_KEY in Settings > Secrets
+# --- 1. SETUP & SECRETS ---
+# Ensure "GROQ_API_KEY" is added to Hugging Face Secrets
 GROQ_API_KEY = os.environ.get("GROQ_API_KEY")
 
-# Initialize LLM (Llama 3 via Groq is ultra-fast)
+# Use Llama 3 for medical reasoning
 llm = ChatGroq(
     temperature=0.1, 
     model_name="llama-3.3-70b-versatile", 
-    groq_api_key=GROQ_API_KEY
+    api_key=GROQ_API_KEY
 )
 
-# Initialize Embeddings (Lightweight & effective for medical text)
+# Best open-source embeddings for medical text
 embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
 
-# --- 2. CORE LOGIC ---
+# --- 2. THE RAG LOGIC ---
 
-def process_and_query(pdf_file, user_question):
+def process_medical_report(pdf_file, user_question):
     if not pdf_file:
-        return "Please upload a medical report first."
+        return "⚠️ Error: Please upload a medical PDF report first."
     
     try:
-        # Step A: Extract Text
+        # Step A: PDF Text Extraction
         doc = fitz.open(pdf_file.name)
         text = ""
         for page in doc:
             text += page.get_text()
         
-        # Step B: Chunking (Medical context needs smaller chunks for precision)
+        if not text.strip():
+            return "⚠️ Error: The PDF seems empty or is an image. Please provide a text-based PDF."
+
+        # Step B: Recursive Chunking
         text_splitter = RecursiveCharacterTextSplitter(
-            chunk_size=600,
-            chunk_overlap=100,
-            separators=["\n\n", "\n", ".", " "]
+            chunk_size=700,
+            chunk_overlap=150
         )
         chunks = text_splitter.split_text(text)
         
-        # Step C: Vector Database (FAISS)
+        # Step C: Temporary Vector Store
         vector_db = FAISS.from_texts(chunks, embeddings)
-        
-        # Step D: Triage & Custom Prompting
-        # This prompt forces the AI to check for urgency and standardize units
-        template = """
-        You are a specialized Medical Cross-Border Assistant. 
-        Context from the patient's report: {context}
-        
-        Guidelines:
-        1. If the user asks for a summary, include: Urgency Level (Low/High), 
-           Key Findings, and standardized units (e.g., convert units if needed).
-        2. Use simple, empathetic language for a migrant/traveler.
-        3. If you see critical values, flag them for immediate doctor consultation.
-        
-        Question: {question}
-        Answer:"""
-        
-        PROMPT = PromptTemplate(template=template, input_variables=["context", "question"])
-        
-        # Step E: RAG Chain
-        chain = RetrievalQA.from_chain_type(
-            llm=llm,
-            chain_type="stuff",
-            retriever=vector_db.as_retriever(search_kwargs={"k": 3}),
-            chain_type_kwargs={"prompt": PROMPT}
+        retriever = vector_db.as_retriever(search_kwargs={"k": 3})
+
+        # Step D: Prompt Engineering (Medical Cross-Border Specialist)
+        medical_template = """
+        You are a Medical Cross-Border AI Agent. Your goal is to help patients understand reports 
+        from different countries.
+
+        CONTEXT FROM REPORT:
+        {context}
+
+        INSTRUCTIONS:
+        1. Summarize findings in simple terms.
+        2. Detect URGENCY: Label as LOW, MEDIUM, or HIGH.
+        3. Convert Units: If you see international units, explain them clearly.
+        4. If a language other than English is requested, provide a high-quality translation.
+        5. CRITICAL: If values are dangerous, tell the user to seek immediate care.
+
+        USER QUESTION: {question}
+        """
+
+        prompt = ChatPromptTemplate.from_template(medical_template)
+
+        # Step E: Modern LCEL Chain (Replaces RetrievalQA)
+        def format_docs(docs):
+            return "\n\n".join(doc.page_content for doc in docs)
+
+        rag_chain = (
+            {"context": retriever | format_docs, "question": RunnablePassthrough()}
+            | prompt
+            | llm
+            | StrOutputParser()
         )
-        
-        response = chain.invoke(user_question)
-        return response["result"]
+
+        # Execute
+        return rag_chain.invoke(user_question)
 
     except Exception as e:
-        return f"Error processing file: {str(e)}"
+        return f"🚨 System Error: {str(e)}"
 
-# --- 3. GRADIO UI DESIGN (Cyber-Luxe Style) ---
+# --- 3. UI DESIGN (Gradio) ---
 
-with gr.Blocks(theme=gr.themes.Soft(primary_hue="blue", secondary_hue="rose")) as demo:
-    gr.Markdown("# 🏥 AI Cross-Border Health Navigator")
-    gr.Markdown("### Standardize, Translate, and Understand your Medical Reports Instantly.")
+# Cyber-Luxe Theme setup
+theme = gr.themes.Soft(
+    primary_hue="cyan",
+    secondary_hue="blue",
+    neutral_hue="slate",
+).set(
+    button_primary_background_fill="*primary_500",
+    button_primary_text_color="white",
+)
+
+with gr.Blocks(theme=theme) as demo:
+    gr.HTML("<h1 style='text-align: center;'>🏥 AI Cross-Border Health Navigator</h1>")
+    gr.Markdown("---")
     
     with gr.Row():
         with gr.Column(scale=1):
-            file_input = gr.File(label="Upload Medical Report (PDF)", file_types=[".pdf"])
-            question_input = gr.Textbox(
-                label="Ask about the report", 
-                placeholder="e.g., 'Summarize this in Urdu' or 'Is my Glucose level high?'"
+            file_upload = gr.File(label="1. Upload Medical PDF", file_types=[".pdf"])
+            chat_query = gr.Textbox(
+                label="2. Ask AI Agent", 
+                placeholder="e.g. 'Summarize and check for urgency' or 'Translate to Urdu'",
+                lines=2
             )
-            submit_btn = gr.Button("Analyze Report", variant="primary")
+            analyze_btn = gr.Button("🚀 Start Analysis", variant="primary")
             
         with gr.Column(scale=2):
-            output_text = gr.Textbox(label="AI Analysis & Insights", interactive=False, lines=15)
-            
-    gr.Examples(
-        examples=[["sample_report.pdf", "Summarize this report and check for medical urgency."]],
-        inputs=[file_input, question_input]
+            output_display = gr.Markdown("### 🔍 AI Insights will appear here...")
+
+    # Action
+    analyze_btn.click(
+        fn=process_medical_report, 
+        inputs=[file_upload, chat_query], 
+        outputs=output_display
     )
 
-    submit_btn.click(fn=process_and_query, inputs=[file_input, question_input], outputs=output_text)
+    gr.Markdown("---")
+    gr.HTML("<p style='text-align: center; color: gray;'>Note: This is an AI prototype for hackathons. Not a substitute for professional medical advice.</p>")
 
 if __name__ == "__main__":
     demo.launch()