Update app.py

app.py

@@ -4,6 +4,13 @@ import numpy as np
 import tempfile
 import os
 
+from langchain_community.document_loaders import UnstructuredImageLoader
+from langchain.prompts import PromptTemplate
+from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace
+
+os.environ["HUGGINGFACEHUB_API_KEY"] = os.getenv("HF")
+os.environ["HF_TOKEN"] = os.getenv("HF")
+
 st.set_page_config(
     page_title="MediAssist - Prescription Analyzer",
     layout="wide",
@@ -43,21 +50,33 @@ uploaded_file = st.file_uploader("📤 Upload Prescription Image (JPG/PNG)", typ
 # Columns for layout
 col1, col2 = st.columns([1, 2])
 
-if uploaded_file:
-    with tempfile.NamedTemporaryFile(delete=False, suffix=".png") as temp_file:
-        temp_file.write(uploaded_file.read())
-        temp_path = temp_file.name
+    dilated_path = orig_path.replace(".png", "_dilated.png")
+    cv2.imwrite(dilated_path, dilated)
+
+    # Load and extract text
+    loader = UnstructuredImageLoader(dilated_path)
+    documents = loader.load()
+    extracted_text = "\n".join([doc.page_content for doc in documents])
 
-    # Step 1: Read and convert to grayscale
-    image = cv2.imread(temp_path)
-    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
+    # Define prompt
+    template = """
+    You are a helpful assistant. Here is a prescription text extracted from an image:
 
-    # Step 2: Inverse Binary Thresholding
-    _, binary_inv = cv2.threshold(gray, 128, 255, cv2.THRESH_BINARY_INV)
+    {prescription_text}
 
-    # Step 3: Dilation with 2 iterations
-    kernel = np.ones((3, 3), np.uint8)
-    dilated = cv2.dilate(binary_inv, kernel, iterations=1)
+    Please summarize the key medicine names and instructions in bullet points.
+    """
+    prompt = PromptTemplate(input_variables=["prescription_text"], template=template)
+
+    model = HuggingFaceEndpoint(
+        repo_id="mistralai/Mistral-7B-Instruct-v0.3",
+        provider="novita",
+        temperature=0.6,
+        max_new_tokens=300,
+        task="conversational"
+    )
+
+    chain = LLMChain(llm=model, prompt=prompt)
 
     with col1:
         st.image(dilated, caption="Preprocessed Prescription", channels="GRAY", use_container_width=True)
@@ -65,20 +84,28 @@ if uploaded_file:
     with col2:
         st.success("✅ Prescription Uploaded & Preprocessed Successfully")
 
-        # Placeholder text (replace with actual OCR + LLM output later)
         st.markdown("### 📜 Extracted Text")
-        st.code("Paracetamol 500mg\nTake 1 tablet twice a day after meals", language='text')
+        st.code(extracted_text)
+    
+        if st.button("🔍 Analyze Text"):
+            with st.spinner("Analyzing..."):
+                response = chain.run(prescription_text=extracted_text)
+            # st.markdown("### LLM Output")
+            st.success(response)
+
 
-        st.markdown("### 🌐 Translated Text")
-        st.code("पेरासिटामोल 500 मिलीग्राम\nभोजन के बाद दिन में दो बार 1 गोली लें", language='text')
+        # st.markdown("### 🌐 Translated Text")
+        # st.code("पेरासिटामोल 500 मिलीग्राम\nभोजन के बाद दिन में दो बार 1 गोली लें", language='text')
 
-        st.markdown("### ⏱️ Tablet Timing & Instructions")
-        st.info("- Morning after breakfast\n- Night after dinner\n- Take with water\n- Do not exceed 2 tablets in 24 hours")
+        # st.markdown("### ⏱️ Tablet Timing & Instructions")
+        # st.info("- Morning after breakfast\n- Night after dinner\n- Take with water\n- Do not exceed 2 tablets in 24 hours")
 
-        st.markdown("### ⚠️ Possible Side Effects")
-        st.warning("- Nausea\n- Dizziness\n- Liver damage (on overdose)")
+        # st.markdown("### ⚠️ Possible Side Effects")
+        # st.warning("- Nausea\n- Dizziness\n- Liver damage (on overdose)")
 
     os.remove(temp_path)
+    os.remove(orig_path)
+    os.remove(dilated_path)
 
 else:
     st.image("https://cdn.dribbble.com/users/285475/screenshots/14711920/media/bd46dc2873f7099e4ef9fd53e6f7f1df.png", width=600)