app.py

import streamlit as st
import torchxrayvision as xrv
import torch
from PIL import Image
import numpy as np
import tensorflow as tf

# Load your pneumonia model
model = tf.keras.models.load_model('model.keras')
class_names = ['Normal', 'Pneumonia']

# Load MobileNetV2 for general image classification
mobilenet = tf.keras.applications.MobileNetV2(weights='imagenet')
imagenet_labels = dict(enumerate(open(tf.keras.utils.get_file(
    'ImageNetLabels.txt',
    'https://storage.googleapis.com/download.tensorflow.org/data/ImageNetLabels.txt'
)).read().splitlines()))

def preprocess_image(image):
    img = image.convert('RGB')
    img = img.resize((224, 224))
    img_array = np.array(img)
    img_array = tf.keras.applications.vgg16.preprocess_input(img_array)
    img_array = np.expand_dims(img_array, axis=0)
    return img_array

st.title('Pneumonia Detection from Chest X-ray')
st.write('Upload a chest X-ray image to predict if it is Normal or Pneumonia.')

uploaded_file = st.file_uploader('Choose an X-ray image...', type=['jpg', 'jpeg', 'png'])

if uploaded_file is not None:
    image = Image.open(uploaded_file)
    st.image(image, caption='Uploaded Image', use_container_width=True)
    st.write('Classifying...')
    img_array = preprocess_image(image)
    prediction = model.predict(img_array)[0][0]
    label = class_names[1] if prediction > 0.5 else class_names[0]
    st.write(f'**Prediction:** {label}')
    st.write(f'Confidence: {prediction:.2f} (closer to 1 = Pneumonia, closer to 0 = Normal)')
    if 0.4 < prediction < 0.6:
        st.warning('This image may not be a valid chest X-ray. Please upload only chest X-ray images.')