app.py

import gradio as gr
import tensorflow as tf
from tensorflow.keras.preprocessing import image
import numpy as np
from PIL import Image

# تحميل الموديل
model = tf.keras.models.load_model("model.h5")

# دالة التنبؤ
def predict(img):
    img = img.resize((228, 228))  # تأكد من نفس المقاس المستخدم في التدريب
    img_array = image.img_to_array(img)
    img_array = np.expand_dims(img_array, axis=0)
    img_array = img_array / 255.0

    prediction = model.predict(img_array)[0][0]
    label = "Stroke" if prediction > 0.5 else "No Stroke"
    confidence = round(float(prediction) * 100, 2) if prediction > 0.5 else round((1 - float(prediction)) * 100, 2)
    return f"{label} ({confidence}%)"

# إنشاء واجهة Gradio
interface = gr.Interface(
    fn=predict,
    inputs=gr.Image(type="pil"),
    outputs="text",
    title="Brain Stroke Detection",
    description="Upload a CT scan image to detect stroke probability."
)

interface.launch()