feat: ✨ Added anonymiser option by default

.pre-commit-config.yaml

@@ -8,7 +8,7 @@ repos:
       - id: check-shebang-scripts-are-executable
       - id: check-added-large-files
         args:
-          - "--maxkb=10000"
+          - "--maxkb=250000"
       # - id: debug-statements # not for process_dcm
       - id: check-yaml
         exclude: mkdocs.yml
@@ -26,18 +26,18 @@ repos:
       - id: check-merge-conflict
       - id: check-docstring-first
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.5.5
+    rev: v0.6.3
     hooks:
       - id: ruff
         args: ["--fix"]
         files: ^hooks
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.5.5
+    rev: v0.6.3
     hooks:
       - id: ruff-format
         files: ^hooks
   - repo: https://github.com/commitizen-tools/commitizen
-    rev: v3.28.0
+    rev: v3.29.0
     hooks:
       - id: commitizen
         stages:
@@ -50,7 +50,7 @@ repos:
           - prettier
   # not for process_dcm
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.11.1
+    rev: v1.11.2
     hooks:
       - id: mypy
         exclude: ^tests/

.talismanrc

@@ -1,12 +1,10 @@
 fileignoreconfig:
-  - filename: .cruft.json
-    checksum: 03b17a0041ffe209156ea08f01b12f2cbb770b4570a6f6678ce8dcae4d945535
   - filename: tests/test_main.py
-    checksum: 3e4ad81306892573518b0e85e38544a978e160e32608229231f8d9f2b2dcd83c
-  - filename: poetry.lock
-    checksum: f4cee8b1c6d6f33b5f34cbbfc093008bcfe7820a26df4225ccea2b3fa5e65e15
+    checksum: efa949f7b639498667939891d29938f01047b7b9a70ffd61a74d47cd127ddbf8
+  - filename: process_dcm/main.py
+    checksum: 3b69ef0a904c0b9e04574fc70e4a35bcf0ed23a75ef9ff99ce6bf80cbaa43551
+  - filename: tests/test_utils.py
+    checksum: 00912e74713f45921e593b5f1f98e23b324ab13bd5c9d44edb24c49a08f325b8
   - filename: process_dcm/utils.py
-    checksum: 4427feb0e00b54eeda3d878b1457be0c0b62f7c39f68e0691442e4ad3988bbf2
-  - filename: .github/workflows/release.yml
-    checksum: 86fbb1303400278bbebbe4523cd4746b4f5e8d1ef036371c8e7a9cbfc0adc254
+    checksum: fdb35056dbd109dbbac6fd5d3668fbff90ace8f46aa37b787eaa2572c02d0aaf
 version: ""

process_dcm/const.py

@@ -3,6 +3,8 @@
 from enum import Enum, Flag, auto, unique
 from typing import cast
 
+RESERVED_CSV = "patient_2_study_id.csv"
+
 
 class ModalityFlag(Flag):
     """A flag representing different modalities in DICOM files."""

process_dcm/main.py

@@ -1,19 +1,19 @@
 """app to procces DCM files."""
 
+import csv
+import os
 from functools import partial
 from multiprocessing import Pool
 
 import typer
 from tqdm import tqdm
 
 from process_dcm import __version__
-from process_dcm.utils import find_dcm_subfolders, process_dcm
+from process_dcm.const import RESERVED_CSV
+from process_dcm.utils import find_dicom_folders_with_base, process_and_save_csv, process_dcm
 
 app = typer.Typer(context_settings={"help_option_names": ["-h", "--help"]})
 
-out_msg = """Output directory for extracted images and metadata. Defaults to: __input_dir__/exported_data\n
-Use absolute path if you want to save the output in a specific location."""
-
 
 def print_version(value: bool) -> None:
     """Print the version of the app."""
@@ -22,30 +22,115 @@ def print_version(value: bool) -> None:
         raise typer.Exit()
 
 
+def process_task(
+    task: tuple[str, str], image_format: str, overwrite: bool, verbose: bool, keep: str, mapping: str
+) -> tuple[str, str]:
+    """Process task."""
+    subfolder, out_dir = task
+    return process_dcm(
+        input_dir=subfolder,
+        output_dir=out_dir,
+        image_format=image_format,
+        overwrite=overwrite,
+        verbose=verbose,
+        keep=keep,
+        mapping=mapping,
+    )
+
+
 @app.command()
 def main(
     input_dir: str = typer.Argument(..., help="Input directory containing subfolders with DICOM files."),
     image_format: str = typer.Option(
         "png", "-f", "--image_format", help="Image format for extracted images (png, jpg, webp). Defaults to: png"
     ),
-    output_dir: str = typer.Option("exported_data", "-o", "--output_dir", help=out_msg),
+    output_dir: str = typer.Option(
+        "exported_data",
+        "-o",
+        "--output_dir",
+        help="Output directory for extracted images and metadata. Defaults to: exported_data",
+    ),
+    relative: bool = typer.Option(
+        False, "-r", "--relative", help="Save extracted data in folders relative to _input_dir_."
+    ),
     n_jobs: int = typer.Option(1, "-j", "--n_jobs", help="Number of parallel jobs. Defaults to: 1"),
+    mapping: str = typer.Option(
+        "",
+        "-m",
+        "--mapping",
+        help=f"Path to CSV containing patient_id to study_id mapping. If not provided and patient_id is not anonymised, a '{RESERVED_CSV}' file will be generated",  # noqa: E501
+    ),
+    keep: str = typer.Option(
+        "",
+        "-k",
+        "--keep",
+        help="Keep the specified fields (p: patient_key, n: names, d: date_of_birth, D: year-only DOB, g: gender)",
+    ),
     overwrite: bool = typer.Option(False, "-w", "--overwrite", help="Overwrite existing images if found."),
     verbose: bool = typer.Option(False, "-v", "--verbose", help="Verbose output."),
     version: bool = typer.Option(
         None, "-V", "--version", callback=print_version, is_eager=True, help="Prints app version"
     ),
 ) -> None:
     """Process DICOM files in subfolders, extract images and metadata using parallel processing."""
-    subfolders = find_dcm_subfolders(input_dir)
+    if mapping == RESERVED_CSV:
+        typer.secho(f"Can't use reserved CSV file name: {RESERVED_CSV}", fg="red")
+        raise typer.Abort()
+    if "p" in keep and mapping:
+        typer.secho(f"WARN:'--mapping' x '--keep p': File , {mapping} it will overwrite patient_id anyway", fg="yellow")
 
-    # Create a partial function with fixed arguments
-    process_dcm_with_args = partial(
-        process_dcm, image_format=image_format, output_dir=output_dir, overwrite=overwrite, verbose=verbose
-    )
+    len_sf, base_dir, subfolders = find_dicom_folders_with_base(input_dir)
+    output_dirs = []
+
+    if relative and os.path.isabs(output_dir):
+        relative = False
+        typer.secho(
+            "WARN: '--relative' x 'absolute --output_dir' are incompatible, absolute 'output_dir' takes precedence",
+            fg="yellow",
+        )
+        output_dirs = [x.replace(base_dir, output_dir) for x in subfolders]
+    elif relative and not os.path.isabs(output_dir):
+        output_dirs = [os.path.join(x, output_dir) for x in subfolders]
+    elif not relative and not os.path.isabs(output_dir):
+        output_dir = os.path.abspath(output_dir)
+        output_dirs = [x.replace(base_dir, output_dir) for x in subfolders]
+
+    tasks = list(zip(subfolders, output_dirs))
 
     with Pool(n_jobs) as pool:
-        results = list(tqdm(pool.imap(process_dcm_with_args, subfolders), total=len(subfolders)))
+        results = list(
+            tqdm(
+                pool.imap(
+                    partial(
+                        process_task,
+                        image_format=image_format,
+                        overwrite=overwrite,
+                        verbose=verbose,
+                        keep=keep,
+                        mapping=mapping,
+                    ),
+                    tasks,
+                ),
+                total=len_sf,
+            )
+        )
+
+    unique_sorted_results = sorted(set(results))  # (study_id, patient_id)
+    dict_res = dict(unique_sorted_results)
+
+    if mapping:
+        with open(mapping) as file:
+            reader = csv.reader(file)
+            mapping_study_ids = set(row[1] for row in reader)
+
+        missing_study_ids = set(result[0] for result in unique_sorted_results) - mapping_study_ids
+
+        for study_id in missing_study_ids:
+            typer.secho(
+                f"Missing map in {mapping}: {dict_res[study_id]} -> {study_id} (<- new hash created)", fg="yellow"
+            )
+    else:
+        process_and_save_csv(unique_sorted_results, RESERVED_CSV)
 
     print(f"Processed {len(results)} DICOM folders.")