app.py

import streamlit as st
import numpy as np
import tensorflow as tf
from PIL import Image
import io

# Load your pre-trained model
model = tf.keras.models.load_model('model.keras')

# Define the image preprocessing function
# Define the image preprocessing function
def preprocess_image(image):
    # Convert image to RGB if it's grayscale
    if image.mode != 'RGB':
        image = image.convert('RGB')
    
    # Resize and preprocess the image
    image = image.resize((224, 224))  # Adjust the size as per your model's requirement
    image_array = np.array(image) / 255.0  # Normalize the image
    image_array = np.expand_dims(image_array, axis=0)  # Add batch dimension
    return image_array

# Define the class labels
class_labels = ['Atelectasis', 'Cardiomegaly', 'Consolidation', 'Edema', 'Effusion', 
                 'Emphysema', 'Fibrosis', 'Infiltration', 'Mass', 
                 'Nodule', 'Pleural_Thickening', 'Pneumothorax']
# class_labels = ["Class1", "Class2", "Class3", "Class4", "Class5"]  # Update with actual class names

# Streamlit app
st.title("Chest X-ray Classification")

# Upload image
uploaded_file = st.file_uploader("Upload a Chest X-ray image...", type=["jpg", "jpeg", "png"])

# Create two columns
col1, col2 = st.columns(2)

if uploaded_file is not None:
    # Read and display the image
    image = Image.open(uploaded_file)
    with col1:
        st.image(image, caption='Uploaded Image', use_column_width=True)
    
    # Preprocess the image
    preprocessed_image = preprocess_image(image)
    
    # Make predictions
    predictions = model.predict(preprocessed_image)[0]
    
    # Get top 3 predictions with probability greater than 0.5
    top_predictions = [(label, prob) for label, prob in zip(class_labels, predictions) if prob > 0.5]
    top_predictions = sorted(top_predictions, key=lambda x: x[1], reverse=True)[:3]
    
    with col2:
        # Display results
        if not top_predictions:
            st.write("No any diseases found with probability greater than 50%.")
        else:
            st.write("Predicted Disease(s):")
            for label, prob in top_predictions:
                st.write(f"{label}: {prob*100:.2f}%")
                percentage = int(prob * 100)
                st.progress(percentage)