Skip to content

Latest commit

 

History

History
180 lines (123 loc) · 4.67 KB

README.md

File metadata and controls

180 lines (123 loc) · 4.67 KB

🌟 Medical Image Processing Tools 🌟

Welcome to the Medical Image Processing Tools repository! This repository provides a collection of tools for processing medical images, specifically DICOM and PNG files, as well as various utility functions.

📌 Table of Contents

🔧 Dependencies

  • 📦 DICOM
  • 📊 Statistics
  • 📄 CSV
  • 📈 DataFrames
  • 📅 Dates
  • 🖼 Images

🖼 DICOM Tools

📖 read_dicom(path_to_file::String)

🔍 Reads a DICOM file.

  • Input:
    • 📂 path_to_file: Path to the DICOM file.
  • Output:
    • 📄 DICOM data.
    • 🆔 SID.
    • 🔑 V_P.
    • 🖼 Image.

📖 get_num_of_patches(img::Matrix{Float32}, lbl::Matrix{Float32}, mask::Matrix{Float32}; thd = 0.35, patch_size = 256)

🔍 Gets the number of patches needed for an image.

  • Input:
    • 🖼 img: The image.
    • 🎯 lbl: The ROI.
    • 🎭 mask: Binary mask representing the breast area.
  • Optional Input:
    • 📏 thd: Threshold of percentage of the white area. Default = 0.35.
    • 📐 patch_size: Size of each patch. Default = 256.
  • Output:
    • 🚫 Number of without-BAC patches.
    • ✅ Number of with-BAC patches.

🛠 Other Tools

📖 histogram_equalization(img; mask=nothing)

🔍 Applies histogram equalization to the image. If the optional mask is provided, it will still process the whole image but only based on the mask area.

  • Input:
    • 🖼 img: Image to be processed.
    • (optional) 🎭 mask: Mask containing only 1 and 0.
  • Output:
    • 🖼 Processed image.

📖 save_to_csv(column_names_, data::Matrix, csv_name::String)

💾 Saves data to a local CSV file.

  • Input:
    • 📋 column_names_: Column names.
    • 📊 data: 2D array of data.
    • 📄 csv_name: Name of the CSV file.

📖 get_last_edit_time(file_path::String)

🕰 Gets the last edit time of a file.

  • Input:
    • 📂 file_path: Path to the file.
  • Output:
    • 📅 Last edit time in Dates format.

📖 search_files_by_ext(root_dir, file_ext)

🔍 Recursively looks for all files in folders and subfolders based on file extension.

  • Input:
    • 📂 root_dir: Root directory.
    • 📄 file_ext: Extension of a file to search for (e.g., ".dcm").
  • Output:
    • 📋 List of found paths.

📖 search_files_by_name(root_dir, file_name)

🔍 Recursively looks for the target file by name.

  • Input:
    • 📂 root_dir: Root directory.
    • 📄 file_name: Name of the target file.
  • Output:
    • 📋 Path to the target file (or an empty string if not found).

📖 binary_search_SID(SIDs, target)

🔍 Looks for a target element in an array (binary search). The input array should be sorted.

  • Input:
    • 📋 SIDs: List containing the target.
    • 🎯 target: The target element.
  • Output:
    • 🔢 Index of the target in the SID list (-1 if not found).

📖 zoom_pixel_values(img; mask=nothing)

🔍 Zooms pixel values of an image to the range [0, 1]. If the optional mask is provided, it will still process the whole image but only based on the mask area.

  • Input:
    • 🖼 img: Input image.
    • (optional) 🎭 mask: Mask containing only 1 and 0.
  • Output:
    • 🖼 Zoomed image.

📖 copy_file(old_path::String, new_path::String)

🔍 Copies a file to a new path.

  • Input:
    • 📂 old_path: Old path of the file.
    • 📂 new_path: New path for the file.

📖 clean_directory(directory::String)

🔍 Deletes everything inside a directory.

  • Input:
    • 📂 directory: Directory to be cleaned.

📖 normalize_img(img; mask=nothing)

🔍 Changes the image, making the mean of all pixel values to 0, and std to 1. If the optional mask is provided, it will still process the whole image but only based on the mask area.

  • Input:
    • 🖼 img: Image to be processed.
    • (optional) 🎭 mask: Mask containing only 1 and 0.
  • Output:
    • 🖼 Processed image.

🎨 PNG Tools

📖 read_png(path_to_file::String)

🔍 Reads a PNG file.

  • Input:
    • 📂 path_to_file: Path to the PNG file.
  • Output:
    • 🖼 Image in Matrix{Float32} format.

📖 get_breast_mask(raw_img::Matrix{Float32})

🎭 Finds the breast area and creates a mask.

  • Input:
    • 🖼 raw_img: The raw image in Matrix{Float32} format.
  • Output:
    • 🎭 Binary mask in Matrix{Float32} format.