# -*- coding: utf-8 -*- """ Created on Mon Aug 5 17:22:37 2024 @author: Aus """ import os import json import nibabel as nib import numpy as np def read_nii_file(path): """ Read a NIfTI file and return the image object and numpy array. """ img = nib.load(path) img_data = img.get_fdata() return img, img_data def get_orientation(img): """ Get the orientation of the NIfTI image. """ return list(nib.aff2axcodes(img.affine)) def print_nii_info(img, img_data, label=False): """ Print detailed information about the NIfTI image or label. """ data_type = "Label" if label else "Image" header = img.header # Commenting out prints for general information # print(f"{data_type} File path: {img.get_filename()}") # print(f"{data_type} Shape: {img_data.shape}") # print(f"{data_type} Affine: \n{img.affine}") # print(f"{data_type} Dimension (dim): {header['dim']}") # print(f"{data_type} Pixel Dimensions (pixdim): {header['pixdim']}") orientation = get_orientation(img) # print(f"{data_type} Orientation: {orientation}") if label: unique_values = np.unique(img_data) min_value = np.min(img_data) max_value = np.max(img_data) num_slices = img_data.shape[2] # Assuming the third dimension is the number of slices slices_with_masks = np.sum(np.any(img_data > 0, axis=(0, 1))) # Commenting out prints for label specific information # print(f"{data_type} Unique values: {unique_values}") # print(f"{data_type} Min value: {min_value}") # print(f"{data_type} Max value: {max_value}") # print(f"{data_type} Number of slices: {num_slices}") # print(f"{data_type} Number of slices with masks: {slices_with_masks}") # print("\n") def save_orientation_to_json(orientation, json_path): """ Save the orientation to a JSON file. """ with open(json_path, 'w') as json_file: json.dump({'orientation': orientation}, json_file) def load_orientation_from_json(json_path): """ Load the orientation from a JSON file. """ with open(json_path, 'r') as json_file: data = json.load(json_file) return data['orientation'] def process_data(images_dir, labels_dir, json_path): """ Process all NIfTI files in the images and labels directories. """ image_files = sorted([f for f in os.listdir(images_dir) if f.endswith(".nii") or f.endswith(".nii.gz")]) label_files = sorted([f for f in os.listdir(labels_dir) if f.endswith(".nii") or f.endswith(".nii.gz")]) first_image = True saved_orientation = None all_same_orientation = True total_slices_with_masks = 0 total_files = len(image_files) for idx, (image_file, label_file) in enumerate(zip(image_files, label_files)): image_path = os.path.join(images_dir, image_file) label_path = os.path.join(labels_dir, label_file) print(f"Processing {image_file} and {label_file}") img, img_data = read_nii_file(image_path) lbl, lbl_data = read_nii_file(label_path) image_orientation = get_orientation(img) label_orientation = get_orientation(lbl) # Save the orientation of the first image to a JSON file if first_image: save_orientation_to_json(image_orientation, json_path) saved_orientation = image_orientation first_image = False print("Orientation saved to JSON") else: saved_orientation = load_orientation_from_json(json_path) print_nii_info(img, img_data) print_nii_info(lbl, lbl_data, label=True) # Check and compare orientations if image_orientation != saved_orientation: print(f"Orientation error: Image file {image_file} has a different orientation: {image_orientation}") all_same_orientation = False if label_orientation != saved_orientation: print(f"Orientation error: Label file {label_file} has a different orientation: {label_orientation}") all_same_orientation = False if image_orientation == saved_orientation and label_orientation == saved_orientation: print("Matching orientation") # Count the total number of slices with masks slices_with_masks = np.sum(np.any(lbl_data > 0, axis=(0, 1))) total_slices_with_masks += slices_with_masks remaining_files = total_files - (idx + 1) print(f"{remaining_files} files remaining") print("-" * 30) if all_same_orientation: print("All images and labels have the same orientation.") print(f"Total number of slices with masks: {total_slices_with_masks}") # Paths to the directories and JSON file labels_dir = r"F:/Task03_Liver/labelsTr" images_dir = r"F:/Task03_Liver/imagesTr" json_path = r"F:/Task03_Liver/orientation.json" print("Processing images and labels...") process_data(images_dir, labels_dir, json_path) ############################################################################################### #testing #load a nifti file and change it's orientation import nibabel as nib import numpy as np def reorient_nifti_file(input_path, output_path, target_orientation): """ Reorient a NIfTI file to the target orientation and save the result. Parameters: input_path (str): Path to the input NIfTI file. output_path (str): Path to save the reoriented NIfTI file. target_orientation (tuple): Target orientation as a tuple, e.g., ('R', 'S', 'A'). """ # Load the NIfTI file img = nib.load(input_path) # Get the current orientation current_orientation = nib.aff2axcodes(img.affine) print(f"Current orientation: {current_orientation}") # Define the transform from current to target orientation transform = nib.orientations.ornt_transform( nib.orientations.axcodes2ornt(current_orientation), nib.orientations.axcodes2ornt(target_orientation) ) # Apply the transformation img_data = img.get_fdata() new_data = nib.orientations.apply_orientation(img_data, transform) # Create a new NIfTI image with the new orientation new_affine = img.affine @ nib.orientations.inv_ornt_aff(transform, img.shape) new_img = nib.Nifti1Image(new_data, new_affine, img.header) # Save the reoriented image nib.save(new_img, output_path) print(f"Reoriented image saved to {output_path}") # Print the orientation of the reoriented file reoriented_orientation = nib.aff2axcodes(new_img.affine) print(f"Reoriented orientation: {reoriented_orientation}") # Example usage input_path = 'E:/LIVER_TUMOR_DETECTION/IMAGES/liver_126.nii' output_path = 'E:/LIVER_TUMOR_DETECTION/IMAGES/liver_126_reoriented.nii' target_orientation = ('R', 'S', 'A') # Desired orientation reorient_nifti_file(input_path, output_path, target_orientation)