CHAT / ANALYZER FIXED

backend.py

@@ -3,7 +3,7 @@ from ast import List
 from fastapi.middleware.cors import CORSMiddleware
 from pydantic import BaseModel
 import io
-import fitz
+#import fitz
 import traceback
 import pandas as pd
 
@@ -85,15 +85,49 @@ class ChatResponse(BaseModel):
 class TextRequest(BaseModel):
     text: str
 
-system_prompt = """ As a highly skilled medical practitioner specializing in image analysis, you are tasked with examining medical images for a renowned hospital. Your expertise is crucial in identifying any anomalies, diseases, or health issues that may be present in the images. Your responsibilities include: 
-1. Detailed Analysis: Thoroughly analyze each image, focusing on identifying any abnormal findings that may indicate underlying medical conditions. 
-2. Finding Report: Document all observed anomalies or signs of disease. Clearly articulate these findings in a structured report format, ensuring accuracy and clarity. Also include any measurement found such as trygliceride, HBa1c, and hdl levels. When presenting any found measurement, relay it in this format: {"findings": "Condition that this measurement could affect only if the value is in a risky rang: masurement type -- value with unit(current range). For example: {"findings": "Diabetes: Hba1c -- 8%(diabetic)", "ldl -- 80mg/dL(optimal)"}"
-    At the beginning of every finding, if needed try to simplify the information given in 3 words.
-3. Checking for Past: If a disease is considered family history or is a disease that the patient recovered from(anything anomoly that is historical), please mention this in severity using this format: "severity" : "severity of the anomoly (Past Anomoly but Still Under Risk)"
-4. Recommendations and Next Steps: Provide detailed recommendations based on your findings. Outline the necessary follow-up actions or additional tests required to confirm diagnoses or assess treatment options. 
-5. Treatment Suggestions: Offer preliminary treatment suggestions or interventions based on the identified conditions, collaborating with the healthcare team to develop comprehensive patient care plans. 
-6. Output Format: Your output should be a JSON array (list) of objects, each describing one disease or medical finding using the structure below: 
-[{"findings": "Description of the first disease or condition.", "severity": "MILD/SEVERE/CRITICAL", "recommendations": ["Follow-up test 1", "Follow-up test 2"], "treatment_suggestions": ["Treatment 1", "Treatment 2"], "home_care_guidance": ["Care tip 1", "Care tip 2"] }, { "findings": "Description of the second disease or condition.", "severity": "MILD/SEVERE/CRITICAL", "recommendations": ["Follow-up test A", "Follow-up test B"], "treatment_suggestions": ["Treatment A", "Treatment B"], "home_care_guidance": ["Care tip A", "Care tip B"] } ] 
+system_prompt = """ You are a highly skilled medical practitioner specializing in medical image and document analysis. You will be given either a medical image or a PDF. 
+
+Your responsibilities are:
+
+1. **Extract Text**: If the input is a PDF or image, first extract all the text content (lab values, notes, measurements, etc.). Do not summarize — keep the extracted text verbatim.  
+
+2. **Detailed Analysis**: Use both the extracted text and the visual features of the image to identify any anomalies, diseases, or health issues.  
+
+3. **Finding Report**: Document all observed anomalies or signs of disease.  
+   - Include any measurements (e.g., triglycerides, HBa1c, HDL) in the format:  
+     `{"findings": "Condition only if risky: measurement type -- value with unit(current range)"}`  
+   - Simplify the finding in **3 words** at the beginning when helpful.  
+
+4. **Checking for Past**: If a disease is family history or previously recovered, mark severity as:  
+   `"severity": "severity of anomaly (Past Anomaly but Still Under Risk)"`
+
+5. **Recommendations and Next Steps**: Provide detailed recommendations (tests, follow-ups, consultations).  
+
+6. **Treatment Suggestions**: Offer preliminary treatments or interventions.  
+
+7. **Output Format**: Always return a JSON object containing both the raw extracted text and the structured analysis, like this:
+
+```json
+{
+  "ocr_text": "<<<FULL VERBATIM TEXT FROM THE PDF/IMAGE>>>",
+  "analysis": [
+    {
+      "findings": "UPPERCASE MAIN CONCERN. Description of the first disease or condition.",
+      "severity": "MILD/SEVERE/CRITICAL",
+      "recommendations": ["Follow-up test 1", "Follow-up test 2"],
+      "treatment_suggestions": ["Treatment 1", "Treatment 2"],
+      "home_care_guidance": ["Care tip 1", "Care tip 2"]
+    },
+    {
+      "findings": "UPPERCASE MAIN CONCERN. Description of the second disease or condition.",
+      "severity": "MILD/SEVERE/CRITICAL",
+      "recommendations": ["Follow-up test A", "Follow-up test B"],
+      "treatment_suggestions": ["Treatment A", "Treatment B"],
+      "home_care_guidance": ["Care tip A", "Care tip B"]
+    }
+  ]
+}
+
 Important Notes: 
     1. Scope of Response: Only respond if the image pertains to a human health issue. 
     2. Clarity of Image: Ensure the image is clear and suitable for accurate analysis. 
@@ -152,7 +186,6 @@ async def analyze_image(image_bytes: bytes, mime_type: str, prompt: Optional[str
         response = await _call_model_blocking(request_inputs, generation_config, safety_settings)
     except Exception as e:
         raise RuntimeError(f"Model call failed: {e}")
-
     text = getattr(response, "text", None)
     if not text and isinstance(response, dict):
         candidates = response.get("candidates") or []
@@ -162,18 +195,25 @@ async def analyze_image(image_bytes: bytes, mime_type: str, prompt: Optional[str
         text = str(response)
 
     clean = re.sub(r"```(?:json)?", "", text).strip()
-
+    print(f"Cleaned text: {clean}")
     try:
         parsed = json.loads(clean)
-        return parsed
+        ocr_text = parsed["ocr_text"]
+        analysis = parsed["analysis"]
+        print(f"Parsed JSON: {parsed}")
+        return analysis,ocr_text
     except json.JSONDecodeError:
         match = re.search(r"(\[.*\]|\{.*\})", clean, re.DOTALL)
         if match:
             try:
-                return json.loads(match.group(1))
+                 
+                parsed =  json.loads(match.group(1)), None
+                ocr_text = parsed["ocr_text"]
+                analysis = parsed["analysis"]
+                return analysis, ocr_text
             except json.JSONDecodeError:
-                return {"raw_found_json": match.group(1)}
-        return {"raw_output": clean}
+                return {"raw_found_json": match.group(1)}, None
+        return {"raw_output": clean}, None
 
 def get_past_reports_from_sqllite(user_id: str):
     try:
@@ -186,6 +226,7 @@ def get_past_reports_from_sqllite(user_id: str):
         history_text = "No past reports found for this user."
     return history_text
 
+
 @app.post("/chat/", response_model=ChatResponse)
 async def chat_endpoint(request: ChatRequest):
     global result
@@ -195,20 +236,20 @@ async def chat_endpoint(request: ChatRequest):
     """
    
     #history_text = get_past_reports_from_firestore(request.user_id)
-    full_document_text = get_past_reports_from_sqllite(request.user_id)
-
-    #full_document_text =  +"\n\n" + "PAST REPORTS:\n" + history_text
-
+    full_document_text = get_past_reports_from_sqllite(request.user_id.strip())
+    
+    full_document_text =  EXTRACTED_TEXT_CACHE+"\n\n" + "PAST REPORTS:\n" + full_document_text
+    print(f"Full document text: {full_document_text}")
     if not full_document_text:
         raise HTTPException(status_code=400, detail="No past reports or current data exists for this user")
     
     
     try:
-        document_text = json.dumps(full_document_text) 
         full_prompt = system_prompt_chat.format(
-            document_text=document_text,
+            document_text=full_document_text,
             user_question=request.question
         )
+        print(f"Full prompt: {full_prompt}")
        
         response = model.generate_content(full_prompt)
         return ChatResponse(answer=response.text)
@@ -223,15 +264,23 @@ async def analyze_endpoint(file: UploadFile = File(...), prompt: str = Form(None
     Returns parsed JSON (or raw model output if JSON couldn't be parsed).
     """
 
-    global result
+    global result,EXTRACTED_TEXT_CACHE
+    
+    filename = file.filename.lower()
+    print(f"Received analyze request for file {filename}")
     contents = await file.read()  # <-- this gets the uploaded file bytes
     mime = file.content_type or "image/png"
-    result = await analyze_image(contents, mime, prompt)
+    
+    #result = await analyze_image(contents, mime, prompt)
     try:
-        result = await analyze_image(contents, mime, prompt)
+        result, ocr_text = await analyze_image(contents, mime, prompt)
+        EXTRACTED_TEXT_CACHE = ocr_text
     except Exception as e:
         raise HTTPException(status_code=500, detail=str(e))
-    return JSONResponse(content={"Detected_Anomolies": result})
+    return JSONResponse(content={
+        "ocr_text": ocr_text,
+        "Detected_Anomolies": result
+    })
 
 @app.post("/analyze_json")
 async def analyze_json(req: AnalyzeRequest):
@@ -250,4 +299,39 @@ def _log_routes():
     print("Mounted routes:")
     for r in app.routes:
         if isinstance(r, APIRoute):
-            print(" ", r.path, r.methods)
+            print(" ", r.path, r.methods)
+
+
+
+def main():
+    """Run the application."""
+    try:
+        logger.info(f"Starting server on 8000")
+        logger.info(f"Debug mode: true")
+
+        if Config.DEBUG:
+            # Use import string for reload mode
+            uvicorn.run(
+                "main:app",
+                host="localhost",
+                port="8000",
+                reload=True,
+                log_level="debug"
+            )
+        else:
+            # Use app instance for production
+            uvicorn.run(
+                app,
+                host="localhost",
+                port="8000",
+                reload=False,
+                log_level="info"
+            )
+
+    except Exception as e:
+        logger.error(f"Failed to start server: {e}")
+        raise
+
+
+if __name__ == "__main__":
+    main()

past_reports.py

@@ -77,7 +77,7 @@ def _safe_parse_json(text):
         return None  # or return s if you prefer to surface the raw text
 
 def _row_to_dict(row: sqlite3.Row) -> dict:
-    return {
+     return {
         "id": row[0],
         "user_id": row[1],
         "report_date": row[2],
@@ -200,4 +200,4 @@ def delete_report(report_id: int):
 if __name__ == "__main__":
     #req = ReportIn(user_id="[email protected]", report_date="2025-09-08", ocr_text="Medical Report - Cancer Patient Name: Carol Davis Age: 55 Gender: Female Clinical History: Recent biopsy confirms breast cancer (invasive ductal carcinoma). No lymph node involvement. PET scan negative for metastasis.", anomalies=[{"findings":"BREAST CANCER(DETECTED AS HISTORICAL CONDITION, BUT STILL UNDER RISK.)","severity":"Severe Risk","recommendations":["Consult a doctor."],"treatment_suggestions":"Consult a specialist: General Practitioner","home_care_guidance":[],"info_link":"https://www.webmd.com/"},{"findings":"CANCER(DETECTED AS HISTORICAL CONDITION, BUT STILL UNDER RISK.)","severity":"Severe Risk","recommendations":["Consult a doctor."],"treatment_suggestions":"Consult a specialist: General Practitioner","home_care_guidance":[],"info_link":"https://www.webmd.com/"},{"findings":"BRAIN CANCER(DETECTED AS HISTORICAL CONDITION, BUT STILL UNDER RISK.)","severity":"Severe Risk","recommendations":["Consult a doctor."],"treatment_suggestions":"Consult a specialist: General Practitioner","home_care_guidance":[],"info_link":"https://www.webmd.com/"}], measurements=[])
     #save_report(req)
-    print(db_fetch_reports(user_id="test.hd@gmail.com"))
+    print(db_fetch_reports(user_id="vineela.local@lowes.com"))

requirements.txt

@@ -32,3 +32,5 @@ numpy==1.25.2
 # --- Cloud / APIs ---
 firebase-admin==5.1.0
 google-generativeai==0.3.1
+
+PyMuPDF==1.22.5 #fitz

web/analyzer.html

@@ -401,7 +401,7 @@
             headers: { "Content-Type": "application/json" },
             body: JSON.stringify({
               question: q,
-              user_id: currentUser ? currentUser.uid : "anonymous",
+              user_id: currentUser ? currentUser.email : "anonymous",
             }),
           });
           if (!res.ok) throw new Error(await res.text());