app.py

import streamlit as st
import pandas as pd
import pickle
import numpy as np
import base64


# Load the background image
def get_base64_of_bin_file(bin_file):
    with open(bin_file, "rb") as f:
        data = f.read()  # Read the file only once
    return base64.b64encode(data).decode()


# Path to the background image
background_image_path = "background2.jpg"  # Replace with your local image path
background_base64 = get_base64_of_bin_file(background_image_path)

# CSS for the full-screen background image
st.markdown(
    f"""
    <style>
    .stApp {{
        background-image: url("data:image/jpeg;base64,{background_base64}");
        background-size: cover;
        background-position: center;
        background-repeat: no-repeat;
        width: 100%;
        height: 100vh;
    }}
    </style>
    """,
    unsafe_allow_html=True
)

# Main content
st.markdown("<h1 style='text-align: center; color: red;'>Heart attack Prediction</h1>", unsafe_allow_html=True)


# Load the dataset
@st.cache_data
def load_dataset():
    return pd.read_csv('heart_disease_health_indicators.csv')  # Update with the path to your dataset

# Cache function to convert DataFrame to CSV
@st.cache_data
def convert_df(df):
    return df.to_csv(index=False).encode("utf-8")

# Load pre-trained model
@st.cache_resource
def load_model():
    with open("model.pkl", "rb") as f:  # Update the model path
        model = pickle.load(f)
    return model

# Load the dataset
df = load_dataset()

# Ensure the dataset has the required columns
required_columns = ['HeartDiseaseorAttack', 'HighBP', 'HighChol', 'CholCheck', 'BMI', 'Smoker', 
                    'Stroke', 'Diabetes', 'PhysActivity', 'Fruits', 'Veggies', 'HvyAlcoholConsump', 
                    'AnyHealthcare', 'NoDocbcCost', 'GenHlth', 'MentHlth', 'PhysHlth', 'DiffWalk', 
                    'Sex', 'Age', 'Education', 'Income']

if not all(col in df.columns for col in required_columns):
    st.error(f"The dataset must contain the following columns: {', '.join(required_columns)}")
else:
    # Add functionality for row selection
    st.markdown("<h3 style='color: red;'>Select a Patient Row for Prediction:</h3>", unsafe_allow_html=True)


    # Option to select a row by index
    st.markdown("<p style='color: red; font-size:16px;'>Select a Row Index:</p>", unsafe_allow_html=True)
    selected_row_index = st.selectbox("", options=range(len(df)), index=0)

    # Add a button below the row selection for heart disease prediction
    predict_button = st.button("Predict Heart attack Risk")

    # If the "Predict" button is clicked
    if predict_button:
        # Row to use for the model
        row_to_use = df.iloc[selected_row_index]

        # Prepare input data for the model (assuming all columns are numeric)
        row_to_use_for_model = row_to_use.drop('HeartDiseaseorAttack')  # Remove target column

        # Load the model
        model = load_model()

        # Check if the number of features matches the model's expectations
        if len(row_to_use_for_model) != model.n_features_in_:
            st.error(f"The model expects {model.n_features_in_} features, but {len(row_to_use_for_model)} were provided.")
        else:
            # Apply the model for heart disease prediction
            prediction = model.predict([row_to_use_for_model])

            # Display the row and the prediction result
            st.markdown("<h3 style='color: red;'>Row selected for heart disease prediction:</h3>", unsafe_allow_html=True)
            st.write(row_to_use)

            # Show the prediction result
            result = "Does Not have Heart Disease or Attack" if prediction[0] == 1 else "Does Not Has Heart Disease or Attack"
            st.markdown(f"<h3 style='color: red;'>Prediction Result: {result}</h3>", unsafe_allow_html=True)


            # Provide option to download the result
            result_df = row_to_use.to_frame().T  # Convert Series to DataFrame
            result_df['Heart Disease Prediction'] = result
            result_csv = convert_df(result_df)
            st.download_button(
                label="Download Prediction Result",
                data=result_csv,
                file_name="Heart_Disease_Prediction_Result.csv",
                mime="text/csv",
            )