Update app.py

app.py

@@ -1,182 +1,141 @@
 import os
-import smtplib
-import mimetypes
-from email.message import EmailMessage
 import gradio as gr
 import tensorflow as tf
 import numpy as np
-from tensorflow.keras.preprocessing import image
+import cv2
+import smtplib
+import ssl
+from email.message import EmailMessage
 from PIL import Image
 from reportlab.lib.pagesizes import letter
 from reportlab.pdfgen import canvas
-from reportlab.lib import colors
-from reportlab.platypus import Table, TableStyle
+from reportlab.lib.utils import ImageReader
 
-# Load the trained model
-model = tf.keras.models.load_model("my_keras_model.h5")
-
-# List of sample images
-sample_images = [f"samples/{img}" for img in os.listdir("samples") if img.endswith((".png", ".jpg", ".jpeg"))]
-
-# Function to process X-ray and generate a PDF report
-def generate_report(name, age, gender, weight, height, address, parents, allergies, cause, email, xray):
-    # Input validation
-    name = name[:50]  
-    address = address[:100]  
-    parents = parents[:50]  
-    cause = ' '.join(cause.split()[:100])  
-
-    image_size = (224, 224)
-
-    def predict_fracture(xray_path):
-        img = Image.open(xray_path).resize(image_size)
-        img_array = image.img_to_array(img) / 255.0
-        img_array = np.expand_dims(img_array, axis=0)
-        prediction = model.predict(img_array)[0][0]
-        return prediction
-
-    prediction = predict_fracture(xray)
-    diagnosed_class = "Normal" if prediction > 0.5 else "Fractured"
-    severity = "Mild" if prediction < 0.3 else "Moderate" if prediction < 0.7 else "Severe"
-
-    treatment_data = [
-        ["Severity Level", "Recommended Treatment", "Recovery Duration"],
-        ["Mild", "Rest, pain relievers, follow-up X-ray", "4-6 weeks"],
-        ["Moderate", "Plaster cast, minor surgery if needed", "6-10 weeks"],
-        ["Severe", "Major surgery, metal implants, physiotherapy", "Several months"]
-    ]
+# Disable GPU to avoid CUDA errors
+os.environ["TF_ENABLE_ONEDNN_OPTS"] = "0"
+os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
 
-    cost_duration_data = [
-        ["Hospital Type", "Estimated Cost", "Recovery Time"],
-        ["Government", f"₹{2000 if severity == 'Mild' else 8000 if severity == 'Moderate' else 20000} - ₹{5000 if severity == 'Mild' else 15000 if severity == 'Moderate' else 50000}", "4-12 weeks"],
-        ["Private", f"₹{10000 if severity == 'Mild' else 30000 if severity == 'Moderate' else 100000}+", "6 weeks - Several months"]
-    ]
-
-    img = Image.open(xray).resize((300, 300))
-    img_path = f"{name}_xray.png"
-    img.save(img_path)
-
-    report_path = f"{name}_fracture_report.pdf"
-    c = canvas.Canvas(report_path, pagesize=letter)
-    
-    c.setFont("Helvetica-Bold", 16)
-    c.drawString(200, 770, "Bone Fracture Detection Report")
-
-    patient_data = [
-        ["Patient Name", name],
-        ["Age", age],
-        ["Gender", gender],
-        ["Weight", f"{weight} kg"],
-        ["Height", f"{height} cm"],
-        ["Address", address],
-        ["Parent's Name", parents],
-        ["Allergies", allergies if allergies else "None"],
-        ["Cause of Injury", cause],
-        ["Diagnosis", diagnosed_class],
-        ["Injury Severity", severity]
-    ]
+# Load the trained model
+model = tf.keras.models.load_model("model.h5")
 
-    def format_table(data):
-        table = Table(data, colWidths=[270, 270])
-        table.setStyle(TableStyle([
-            ('BACKGROUND', (0, 0), (-1, 0), colors.darkblue),
-            ('TEXTCOLOR', (0, 0), (-1, 0), colors.whitesmoke),
-            ('ALIGN', (0, 0), (-1, -1), 'CENTER'),
-            ('FONTNAME', (0, 0), (-1, 0), 'Helvetica-Bold'),
-            ('BOTTOMPADDING', (0, 0), (-1, 0), 12),
-            ('GRID', (0, 0), (-1, -1), 1, colors.black),
-            ('VALIGN', (0, 0), (-1, -1), 'MIDDLE')
-        ]))
-        return table
-
-    patient_table = format_table(patient_data)
-    patient_table.wrapOn(c, 480, 500)
-    patient_table.drawOn(c, 50, 620)
-
-    c.drawInlineImage(img_path, 170, 320, width=250, height=250)
-    c.setFont("Helvetica-Bold", 12)
-    c.drawString(250, 290, f"Fractured: {'Yes' if diagnosed_class == 'Fractured' else 'No'}")
-
-    treatment_table = format_table(treatment_data)
-    treatment_table.wrapOn(c, 480, 200)
-    treatment_table.drawOn(c, 50, 200)
-
-    cost_table = format_table(cost_duration_data)
-    cost_table.wrapOn(c, 480, 150)
-    cost_table.drawOn(c, 50, 80)
+# Email sender credentials (Set your own credentials here)
+SENDER_EMAIL = "[email protected]"
+SENDER_PASSWORD = "your_email_password"
 
-    c.save()
+def send_email(receiver_email, file_path):
+    """Function to send an email with the generated PDF attached"""
+    msg = EmailMessage()
+    msg["Subject"] = "Bone Fracture Patient Report"
+    msg["From"] = SENDER_EMAIL
+    msg["To"] = receiver_email
+    msg.set_content("Please find the attached bone fracture report.")
 
-    send_email(email, report_path)  
+    # Attach PDF file
+    with open(file_path, "rb") as f:
+        file_data = f.read()
+        msg.add_attachment(file_data, maintype="application", subtype="pdf", filename="Fracture_Report.pdf")
 
-    return report_path  
+    # Send email
+    context = ssl.create_default_context()
+    with smtplib.SMTP_SSL("smtp.gmail.com", 465, context=context) as server:
+        server.login(SENDER_EMAIL, SENDER_PASSWORD)
+        server.send_message(msg)
 
-# Function to send email
-def send_email(receiver_email, attachment_path):
-    sender_email = "yourhospital@example.com"  
-    sender_password = "yourpassword"  
+def preprocess_image(image):
+    """Preprocess the image for model prediction"""
+    image = cv2.cvtColor(np.array(image), cv2.COLOR_RGB2GRAY)
+    image = cv2.resize(image, (224, 224))
+    image = np.expand_dims(image, axis=-1)
+    image = np.expand_dims(image, axis=0) / 255.0
+    return image
+
+def generate_pdf(name, age, gender, weight, height, allergies, injury_cause, address, parent_name, image, email):
+    """Generate a PDF report"""
+    file_path = "Fracture_Report.pdf"
+    c = canvas.Canvas(file_path, pagesize=letter)
+    width, height = letter
+
+    # Title
+    c.setFont("Helvetica-Bold", 16)
+    c.drawCentredString(width / 2, height - 50, "Bone Fracture Patient Report")
+
+    # Patient Info Table
+    c.setFont("Helvetica", 12)
+    data = [
+        ["Patient Name:", name[:50]],
+        ["Age:", age],
+        ["Gender:", gender],
+        ["Weight (kg):", weight],
+        ["Height (cm):", height],
+        ["Allergies:", allergies[:100]],
+        ["Injury Cause:", " ".join(injury_cause.split()[:100])],  # Limit to 100 words
+        ["Address:", address[:100]],
+        ["Parent/Guardian Name:", parent_name[:50]],
+    ]
 
-    msg = EmailMessage()
-    msg["Subject"] = "Your Bone Fracture Report"
-    msg["From"] = sender_email
-    msg["To"] = receiver_email
-    msg.set_content("Please find attached your bone fracture diagnosis report.")
+    x_start, y_start = 50, height - 100
+    line_spacing = 20
 
-    mime_type, _ = mimetypes.guess_type(attachment_path)
-    mime_type = mime_type or "application/octet-stream"
+    for row in data:
+        c.drawString(x_start, y_start, f"{row[0]} {row[1]}")
+        y_start -= line_spacing
 
-    with open(attachment_path, "rb") as attachment:
-        msg.add_attachment(attachment.read(), maintype=mime_type.split("/")[0], subtype=mime_type.split("/")[1], filename=os.path.basename(attachment_path))
+    # Add X-ray image
+    if image:
+        img = Image.open(image)
+        img.thumbnail((250, 250))
+        img_path = "temp_image.jpg"
+        img.save(img_path)
+        c.drawImage(ImageReader(img_path), width / 2 - 125, y_start - 250, 250, 250)
 
-    with smtplib.SMTP_SSL("smtp.example.com", 465) as server:
-        server.login(sender_email, sender_password)
-        server.send_message(msg)
+    # Close and save the PDF
+    c.save()
 
-# Function to select a sample image
-def use_sample_image(sample_image_path):
-    return sample_image_path  
-
-# Define Gradio Interface
-with gr.Blocks() as app:
-    gr.Markdown("## Bone Fracture Detection System")
-    
-    with gr.Row():
-        name = gr.Textbox(label="Patient Name (Max 50 chars)", max_chars=50)
-        age = gr.Number(label="Age")
-        gender = gr.Radio(["Male", "Female", "Other"], label="Gender")
-    
-    with gr.Row():
-        weight = gr.Number(label="Weight (kg)")
-        height = gr.Number(label="Height (cm)")
-    
-    with gr.Row():
-        address = gr.Textbox(label="Address (Max 100 chars)", max_chars=100)
-        parents = gr.Textbox(label="Parent's Name (Max 50 chars)", max_chars=50)
-    
-    with gr.Row():
-        allergies = gr.Textbox(label="Allergies (if any)")
-        cause = gr.Textbox(label="Cause of Injury (Max 100 words)")
-
-    with gr.Row():
-        email = gr.Textbox(label="Patient's Email (To receive report)", type="email")
-    
-    with gr.Row():
-        xray = gr.Image(type="filepath", label="Upload X-ray Image")
-    
-    with gr.Row():
-        sample_selector = gr.Dropdown(choices=sample_images, label="Use Sample Image")
-        select_button = gr.Button("Load Sample Image")
-
-    submit_button = gr.Button("Generate Report")
-    output_file = gr.File(label="Download Report")
-
-    select_button.click(use_sample_image, inputs=[sample_selector], outputs=[xray])
-
-    submit_button.click(
-        generate_report,
-        inputs=[name, age, gender, weight, height, address, parents, allergies, cause, email, xray],
-        outputs=[output_file],
-    )
+    # Send email
+    if email:
+        send_email(email, file_path)
+
+    return file_path
+
+def predict_and_generate_report(name, age, gender, weight, height, allergies, injury_cause, address, parent_name, image, email):
+    """Make a prediction and generate a report"""
+    if image is None:
+        return "Please upload an X-ray image."
+
+    # Preprocess and make a prediction
+    processed_image = preprocess_image(image)
+    prediction = model.predict(processed_image)
+    confidence = float(prediction[0][0]) * 100
+    fracture_status = "Yes" if confidence > 50 else "No"
+
+    # Generate PDF report
+    pdf_path = generate_pdf(name, age, gender, weight, height, allergies, injury_cause, address, parent_name, image, email)
+
+    return f"Fractured: {fracture_status} (Confidence: {confidence:.2f}%)", pdf_path
+
+# Define the Gradio Interface
+iface = gr.Interface(
+    fn=predict_and_generate_report,
+    inputs=[
+        gr.Textbox(label="Patient Name (Max 50 chars)"),
+        gr.Number(label="Age", precision=0),
+        gr.Radio(label="Gender", choices=["Male", "Female", "Other"]),
+        gr.Number(label="Weight (kg)"),
+        gr.Number(label="Height (cm)"),
+        gr.Textbox(label="Allergies (Max 100 chars)"),
+        gr.Textbox(label="Cause of Injury (Max 100 words)"),
+        gr.Textbox(label="Address (Max 100 chars)"),
+        gr.Textbox(label="Parent/Guardian Name (Max 50 chars)"),
+        gr.Image(type="pil", label="Upload X-ray Image"),
+        gr.Textbox(label="Email Address (for Report)"),
+    ],
+    outputs=[
+        gr.Textbox(label="Fracture Prediction"),
+        gr.File(label="Download Report"),
+    ],
+    title="Bone Fracture Detection System",
+    description="Upload an X-ray image, enter patient details, and generate a fracture report."
+)
 
 if __name__ == "__main__":
-    app.launch()
+    iface.launch()