app.py

import pytesseract as pyt
import cv2
import re
import gradio as gr

# Function to extract text from image and identify relevant details
def extract_patient_info(image):
    # Convert the image to a numpy array
    img = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)  # Gradio loads the image as RGB, OpenCV uses BGR
    text = pyt.image_to_string(img)  # Use Tesseract to extract text from the image
    
    patient_info = {
        "Name": None,
        "Age": None,
        "Gender": None,
        "Phone Number": None
    }
    
    # Debug: Print the extracted OCR text for troubleshooting
    print("OCR Text:\n", text)
    
    # Extract Patient Name using regex (Look for 'Patient Name:' and capture the name)
    name_match = re.search(r"Name[:\s]*([A-Za-z\s]+)", text)
    if name_match:
        patient_info["Name"] = name_match.group(1)
    
    # Extract Age using regex (Look for 'Age' and capture the value)
    age_match = re.search(r"Age\s*[:\-]?\s*(\d{1,2})", text)
    if age_match:
        patient_info["Age"] = age_match.group(1)
    
    # Extract Gender using regex (Look for 'Gender' and capture the gender)
    gender_match = re.search(r"Gender\s*[:\-]?\s*(Male|Female)", text, re.IGNORECASE)
    if gender_match:
        patient_info["Gender"] = gender_match.group(1)
    
    # Extract Phone Number using regex (Look for a 10-digit number)
    phone_match = re.search(r"\b\d{10}\b", text)
    if phone_match:
        patient_info["Phone Number"] = phone_match.group(0)
    
    return patient_info["Name"], patient_info["Age"], patient_info["Gender"], patient_info["Phone Number"]

# Gradio interface setup
with gr.Blocks() as demo:
    gr.Markdown("### Patient Information Extraction from Image")
    
    # Image upload component
    image_input = gr.Image(type="numpy", label="Upload Image")
    
    # Output textboxes to display the extracted information
    name_output = gr.Textbox(label="Name")
    age_output = gr.Textbox(label="Age")
    gender_output = gr.Textbox(label="Gender")
    phone_output = gr.Textbox(label="Phone Number")
    
    # Button to trigger image processing and text extraction
    process_button = gr.Button("Process Image")
    
    # When the button is clicked, process the image and show results in textboxes
    process_button.click(fn=extract_patient_info, inputs=image_input, outputs=[name_output, age_output, gender_output, phone_output])

# Launch the Gradio app
if __name__ == "__main__":
    demo.launch()