AQC_dataset / rename_acdc_specific.py
maxmo2009's picture
Upload folder using huggingface_hub
ea4eb5a verified
#!/usr/bin/env python3
"""
Specific renaming script for ACDC dataset (adapted for HuggingFace download structure)
Input: flat Images/ and Masks/ folders with patient###_frameXX[_gt].nii.gz
Output: per-patient subdirectories with renamed files:
- patient###_frameXX.nii.gz -> lvsa_SR_ED.nii.gz (lower frame) or lvsa_SR_ES.nii.gz (higher frame)
- patient###_frameXX_gt.nii.gz -> seg_lvsa_SR_ED.nii.gz (lower frame) or seg_lvsa_SR_ES.nii.gz (higher frame)
"""
import re
import shutil
from pathlib import Path
from collections import defaultdict
import argparse
def rename_acdc_files(source_dir: str, output_dir: str, dry_run: bool = True):
"""
Reorganise flat Images/Masks folders into per-patient directories with renamed files.
Args:
source_dir: Directory containing Images/ and Masks/ subdirectories
output_dir: Directory where per-patient folders will be created
dry_run: If True, only show what would be done
"""
source_path = Path(source_dir)
output_path = Path(output_dir)
images_dir = source_path / "Images"
masks_dir = source_path / "Masks"
if not images_dir.exists() or not masks_dir.exists():
print(f"Error: Expected Images/ and Masks/ subdirectories in {source_path}")
return
print(f"\n{'DRY RUN MODE' if dry_run else 'EXECUTING'}")
print("=" * 50)
# Group files by patient
patients = defaultdict(dict) # patient_id -> {frame_num: {'img': path, 'gt': path}}
for img_file in sorted(images_dir.glob("patient*_frame*.nii.gz")):
match = re.match(r'(patient\d+)_frame(\d+)\.nii\.gz', img_file.name)
if match:
pid, fnum = match.group(1), int(match.group(2))
patients[pid].setdefault(fnum, {})['img'] = img_file
for gt_file in sorted(masks_dir.glob("patient*_frame*_gt.nii.gz")):
match = re.match(r'(patient\d+)_frame(\d+)_gt\.nii\.gz', gt_file.name)
if match:
pid, fnum = match.group(1), int(match.group(2))
patients[pid].setdefault(fnum, {})['gt'] = gt_file
for pid in sorted(patients):
frame_files = patients[pid]
frame_numbers = sorted(frame_files.keys())
print(f"\nProcessing: {pid}")
if len(frame_numbers) < 2:
print(f" Warning: Only {len(frame_numbers)} frame found, expected 2")
continue
ed_frame = frame_numbers[0]
es_frame = frame_numbers[-1]
print(f" ED frame: {ed_frame}, ES frame: {es_frame}")
patient_out = output_path / pid
if not dry_run:
patient_out.mkdir(parents=True, exist_ok=True)
rename_map = {
(ed_frame, 'img'): 'lvsa_SR_ED.nii.gz',
(ed_frame, 'gt'): 'seg_lvsa_SR_ED.nii.gz',
(es_frame, 'img'): 'lvsa_SR_ES.nii.gz',
(es_frame, 'gt'): 'seg_lvsa_SR_ES.nii.gz',
}
for (fnum, ftype), new_name in rename_map.items():
if ftype in frame_files.get(fnum, {}):
src = frame_files[fnum][ftype]
dst = patient_out / new_name
if dry_run:
print(f" Would copy: {src.name} -> {pid}/{new_name}")
else:
shutil.copy2(src, dst)
print(f" Copied: {src.name} -> {pid}/{new_name}")
print("\n" + "=" * 50)
if dry_run:
print("DRY RUN COMPLETE - No files were actually copied")
print("To execute, run with --execute flag")
else:
print(f"COMPLETE - Output at {output_path}")
def main():
parser = argparse.ArgumentParser(description='Reorganise and rename ACDC dataset files')
parser.add_argument('--source', type=str,
default='/mnt/storage/home/ym1413/QC_data_preprocessing/raw_dataset/ACDC',
help='Source directory containing Images/ and Masks/')
parser.add_argument('--output', type=str,
default='/mnt/storage/home/ym1413/QC_data_preprocessing/sabotaged_dataset',
help='Output directory for per-patient folders')
parser.add_argument('--dry-run', action='store_true', default=True,
help='Show what would be done without executing (default)')
parser.add_argument('--execute', action='store_true',
help='Actually perform the copy and rename')
args = parser.parse_args()
if args.execute:
args.dry_run = False
rename_acdc_files(args.source, args.output, dry_run=args.dry_run)
if __name__ == "__main__":
main()