如何用Python自动化在各子文件夹运行Slicer DICOM处理脚本？

Problem Description #

• I have the following file structure: A root folder W001-W100 containing approximately 1000 subfolders named W001 to W100, each with around 500 CT DICOM files.
• I'm using Slicer and Python to generate surface meshes, and I already have a processing script, but it requires manual modification of the subfolder path and .ply save path each time.
• Since there are a huge number of subfolders, I want to implement automated processing via Python. I'm more familiar with R and don't know much about Python, so I need detailed automation steps and code modification plans that fit my needs.

Answer #

Step-by-Step Automation Plan #

1. Set Up Path Variables: Define the root directory of your DICOM folders and the output directory for .ply files (no manual changes needed after this).
2. Iterate Through Subfolders: Use Python's built-in os module to automatically list and loop through all subfolders in the root directory.
3. Integrate with Your Existing Slicer Script: Replace hardcoded paths in your script with dynamic paths generated during the loop.
4. Error Handling (Optional but Recommended): Add basic checks to skip non-DICOM folders or log failures.

Modified Code with Annotations #

Here's the adapted Python code that works with your Slicer workflow. I've added detailed comments since you're less familiar with Python:

# Import Python's built-in module for file system operations
import os

# --------------------------
# Configure these two paths once
# --------------------------
# Path to your root folder (W001-W100)
ROOT_DICOM_FOLDER = "C:/path/to/your/W001-W100"
# Path to the folder where you want to save all .ply files
OUTPUT_PLY_FOLDER = "C:/path/to/your/ply_outputs"

# --------------------------
# Create output folder if it doesn't exist
# --------------------------
if not os.path.exists(OUTPUT_PLY_FOLDER):
    os.makedirs(OUTPUT_PLY_FOLDER)
    print(f"Created output folder: {OUTPUT_PLY_FOLDER}")

# --------------------------
# Loop through every subfolder in the root directory
# --------------------------
for subfolder_name in os.listdir(ROOT_DICOM_FOLDER):
    # Get full path to the current subfolder
    subfolder_path = os.path.join(ROOT_DICOM_FOLDER, subfolder_name)
    
    # Skip if it's not a folder (in case there are files in the root directory)
    if not os.path.isdir(subfolder_path):
        continue
    
    # --------------------------
    # Define the save path for the .ply file
    # Name it after the subfolder (e.g., W001_mesh.ply)
    # --------------------------
    ply_save_path = os.path.join(OUTPUT_PLY_FOLDER, f"{subfolder_name}_mesh.ply")
    
    # --------------------------
    # Insert your existing Slicer processing code here
    # Replace any hardcoded paths with subfolder_path and ply_save_path
    # Example of how your existing code might be modified:
    # --------------------------
    print(f"Processing folder: {subfolder_name}")
    
    # 1. Load DICOM series from subfolder_path (your existing code)
    # slicer.util.loadVolume(subfolder_path)
    
    # 2. Run surface mesh generation (your existing code)
    # ... (your mesh creation steps)
    
    # 3. Save the mesh to ply_save_path (your existing code)
    # slicer.util.saveNode(meshNode, ply_save_path)
    
    print(f"Saved mesh to: {ply_save_path}\n")

Key Explanations for R Users #

• os.listdir(): Equivalent to R's list.dirs() or list.files() – it gets all items in a directory.
• os.path.join(): Safely combines folder paths (avoids issues with slashes on Windows/macOS, like R's file.path()).
• os.path.isdir(): Checks if an item is a folder, similar to R's file.info()$isdir.
• The f"{variable}_text" syntax: Python's string interpolation, like R's paste0(variable, "_text").

Important Notes #

• Test with a small subset first: Copy 2-3 subfolders to a test root directory and run the script to make sure it works before processing all 1000 folders.
• Slicer's Python environment: This code runs directly in Slicer's built-in Python console (no need for external Python installations).
• Logging: If you want to track progress/failures, add code to write to a log file (e.g., with open("processing_log.txt", "a") as f: f.write(f"Processed {subfolder_name}\n")).

内容的提问来源于stack exchange，提问作者antecessor