config.py

import os
from dotenv import load_dotenv

load_dotenv()

# Set HF token to avoid rate limit warnings
os.environ["HUGGINGFACE_HUB_TOKEN"] = os.getenv("HF_TOKEN", "")# -----------------------------------------------
# Medical Report Analyzer — Central Configuration
# All project settings live here
# -----------------------------------------------

# -----------------------------------------------
# HuggingFace Dataset Settings
# -----------------------------------------------
DATASET_NAME = "kamruzzaman-asif/Diseases_Dataset"

# Splits we use and why
DATASET_SPLITS = {
    "IndianServers":  "India-specific diseases",
    "QuyenAnh":       "Has treatment data",
    "symptom2disease": "Natural language symptoms",
    "itachi9604":     "Additional disease entries",
}

# -----------------------------------------------
# File Paths
# -----------------------------------------------
MEDICAL_KNOWLEDGE_FILE = "data/medical_knowledge.txt"
FAISS_INDEX_PATH       = "faiss_index"
REPORTS_OUTPUT_PATH    = "generated_reports"

# -----------------------------------------------
# RAG Settings
# -----------------------------------------------
CHUNK_SIZE    = 500    # characters per chunk
CHUNK_OVERLAP = 100    # overlap between chunks
TOP_K_RESULTS = 5      # chunks retrieved per query

# Embedding model — runs locally, no API needed
EMBEDDING_MODEL = "sentence-transformers/all-MiniLM-L6-v2"

# -----------------------------------------------
# Gemini Settings
# -----------------------------------------------
GEMINI_MODEL = "gemini-2.5-flash"

# -----------------------------------------------
# Supported Document Types
# -----------------------------------------------
SUPPORTED_FORMATS = [".pdf", ".jpg", ".jpeg", ".png"]

# -----------------------------------------------
# Language Settings
# -----------------------------------------------
SUPPORTED_LANGUAGES = {
    "English": "en",
    "Hindi":   "hi",
}

# -----------------------------------------------
# Risk Levels
# -----------------------------------------------
RISK_LEVELS = {
    "LOW":      "🟢 Low Risk",
    "MEDIUM":   "🟡 Medium Risk",
    "HIGH":     "🔴 High Risk",
    "CRITICAL": "🚨 Critical — Seek Immediate Help",
}

# -----------------------------------------------
# Disclaimer — added to every response
# -----------------------------------------------
DISCLAIMER = """
⚠️ MEDICAL DISCLAIMER:
This analysis is AI-generated and for informational purposes only.
It does not constitute medical advice or diagnosis.
Always consult a qualified doctor before making health decisions.
Emergency: Call 112 | AIIMS Helpline: 011-26588500
Free Health Helpline: 104
"""

IMAGING_DISCLAIMER = """
⚠️ IMAGING DISCLAIMER:
X-ray, MRI and CT scan analysis by AI is for general guidance only.
AI cannot replace a qualified radiologist for imaging interpretation.
Always get your scans reviewed by a certified radiologist.
"""