import streamlit as st
import joblib
import pandas as pd

# Custom CSS
st.markdown(
    """
    <style>
    .stApp {
        background: url('https://mandayahospitalgroup.com/wp-content/uploads/2024/05/diabetes.jpg') no-repeat center center fixed;
        background-size: cover;
    }
    .stApp h1 {
        background-color: rgba(0, 0, 128, 0.7); 
        color: #ffffff; 
        padding: 10px;
        border-radius: 5px;
        font-size: 2.2em; 
        text-align: center;
        white-space: nowrap;
        overflow: hidden;
        text-overflow: ellipsis;
        max-width: 100%;
        margin: 0 auto;
    }
    .stButton>button {
        background-color: #4CAF50;
        color: white;
        font-size: 1.2em;
        border-radius: 10px;
        padding: 10px 24px;
        border: none;
    }
    .stButton {
        display: flex;
        justify-content: center;
    }
    .positive-result {
        background-color: rgba(0, 128, 0, 0.8);
        color: white;
        font-size: 1.5em;
        padding: 20px;
        border-radius: 12px;
        margin-top: 25px;
        text-align: center;
        box-shadow: 0 4px 10px rgba(0, 0, 0, 0.2);
    }
    .negative-result {
        background-color: rgba(220, 20, 60, 0.85);
        color: white;
        font-size: 1.5em;
        padding: 20px;
        border-radius: 12px;
        margin-top: 25px;
        text-align: center;
        box-shadow: 0 4px 10px rgba(0, 0, 0, 0.2);
    }
    </style>
    """,
    unsafe_allow_html=True
)

# Load models
model = joblib.load("lr_model.joblib")
encoder = joblib.load("encoder_d.joblib")
scaler = joblib.load("scaler.joblib")

# Title
st.title("🔍 Smart Diabetes Risk Assessment System")
st.write("Provide the following details to assess risk factors for diabetes.")

# Input columns
col1, col2, col3 = st.columns(3)

with col1:
    bmi = st.number_input("Body Mass Index (BMI):", 10.0, 50.0, step=0.1)
    family_history = st.selectbox("Family History of Diabetes:", encoder["Family_History"].classes_)
    family_history = encoder["Family_History"].transform([family_history])[0]
    fasting_blood_sugar = st.number_input("Fasting Blood Sugar (mg/dL):", 50, 300, step=1)
    hba1c = st.number_input("HBA1C (%):", 3.0, 15.0, step=0.1)
    age = st.number_input("Age (years):", 1, 100, step=1)

with col2:
    physical_activity = st.selectbox("Physical Activity Level:", encoder["Physical_Activity"].classes_)
    physical_activity = encoder["Physical_Activity"].transform([physical_activity])[0]
    postprandial_blood_sugar = st.number_input("Postprandial Blood Sugar (mg/dL):", 50, 400, step=1)
    waist_hip_ratio = st.number_input("Waist-to-Hip Ratio:", 0.5, 2.0, step=0.01)
    vitamin_d_level = st.number_input("Vitamin D Level (ng/mL):", 5.0, 100.0, step=0.1)

with col3:
    diet_type = st.selectbox("Diet Type:", encoder["Diet_Type"].classes_)
    diet_type = encoder["Diet_Type"].transform([diet_type])[0]
    stress_level = st.selectbox("Stress Level:", encoder["Stress_Level"].classes_)
    stress_level = encoder["Stress_Level"].transform([stress_level])[0]
    glucose_tolerance = st.number_input("Glucose Tolerance Test Result (mg/dL):", 50, 300, step=1)
    c_protein_level = st.number_input("C-Reactive Protein Level (mg/L):", 0.1, 20.0, step=0.1)
    cholesterol_level = st.number_input("Cholesterol Level (mg/dL):", 100, 400, step=1)

# Prediction logic
values = [bmi, family_history, physical_activity, diet_type, stress_level, fasting_blood_sugar,
          postprandial_blood_sugar, hba1c, waist_hip_ratio, glucose_tolerance, age,
          vitamin_d_level, c_protein_level, cholesterol_level]

if st.button("Submit"):
    scaled_values = scaler.transform([values])
    prediction = model.predict(scaled_values)

    if prediction[0] == 1:
        st.markdown(
            '<div class="negative-result">⚠️ <strong>Risk Alert:</strong> Based on the input data, there is a <strong>significant likelihood</strong> of diabetes. Please consult a healthcare provider for further evaluation.</div>',
            unsafe_allow_html=True
        )
    else:
        st.markdown(
            '<div class="positive-result">✅ <strong>Good News:</strong> Based on the input data, there appears to be <strong>no immediate risk</strong> of diabetes. Keep maintaining a healthy lifestyle!</div>',
            unsafe_allow_html=True
        )