Add DICOM to ImageScope XML converter

sectra_dicom_to_imagescope.py

@@ -0,0 +1,112 @@
+#!/usr/bin/env python3
+"""Convert DICOM annotations to Aperio ImageScope XML.
+
+This script reads annotation data from a DICOM file and
+writes a minimal ImageScope-compatible XML file. Only basic
+annotation types (polylines/polygons and circles) are handled.
+"""
+
+import argparse
+import os
+import xml.etree.ElementTree as ET
+from typing import Iterable, List, Tuple
+
+import pydicom
+from pydicom.dataset import Dataset
+
+
+def _parse_float(val: object) -> float:
+    """Safely parse a value as float."""
+    try:
+        return float(val)
+    except Exception:
+        return 0.0
+
+
+def _extract_coordinates(data: object) -> List[Tuple[float, float]]:
+    """Extract (x, y) pairs from a DICOM GraphicData item."""
+    if data is None:
+        return []
+    # Convert to list of strings
+    if isinstance(data, (list, tuple)):
+        parts = list(data)
+    else:
+        text = str(data)
+        parts = text.split("\\")
+    coords: List[Tuple[float, float]] = []
+    for i in range(0, len(parts), 2):
+        if i + 1 < len(parts):
+            x = _parse_float(parts[i])
+            y = _parse_float(parts[i + 1])
+            coords.append((x, y))
+    return coords
+
+
+def _extract_graphics(seq: Dataset) -> Iterable[List[Tuple[float, float]]]:
+    """Yield coordinate lists for graphic objects in a sequence."""
+    if hasattr(seq, "GraphicObjectSequence"):
+        for obj in seq.GraphicObjectSequence:
+            coords = _extract_coordinates(getattr(obj, "GraphicData", None))
+            if coords:
+                yield coords
+
+
+def dicom_to_imagescope_xml(ds: Dataset) -> ET.Element:
+    """Convert a DICOM dataset to an ImageScope Annotations element."""
+    root = ET.Element("Annotations")
+    ann_id = 0
+    region_id = 0
+    for seq in getattr(ds, "GraphicAnnotationSequence", []):
+        polygons = list(_extract_graphics(seq))
+        if not polygons:
+            continue
+        ann_id += 1
+        ann = ET.SubElement(
+            root,
+            "Annotation",
+            {
+                "Id": str(ann_id),
+                "Name": f"Annotation {ann_id}",
+                "ReadOnly": "0",
+            },
+        )
+        regions = ET.SubElement(ann, "Regions")
+        for poly in polygons:
+            region_id += 1
+            reg = ET.SubElement(
+                regions,
+                "Region",
+                {
+                    "Id": str(region_id),
+                    "Type": "0",
+                    "LineColor": "65280",
+                },
+            )
+            verts = ET.SubElement(reg, "Vertices")
+            for x, y in poly:
+                ET.SubElement(verts, "Vertex", {"X": str(x), "Y": str(y)})
+    return root
+
+
+def dicom_file_to_imagescope(dicom_path: str, output: str | None = None) -> None:
+    ds = pydicom.dcmread(dicom_path)
+    root = dicom_to_imagescope_xml(ds)
+    xml_bytes = ET.tostring(root, encoding="utf-8")
+    if output is None:
+        output = os.path.splitext(dicom_path)[0] + "_aperio.xml"
+    with open(output, "wb") as f:
+        f.write(xml_bytes)
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(
+        description="Convert DICOM annotations to Aperio ImageScope XML"
+    )
+    parser.add_argument("-i", "--input", required=True, help="Input DICOM file")
+    parser.add_argument("-o", "--output", help="Output XML file")
+    args = parser.parse_args()
+    dicom_file_to_imagescope(args.input, args.output)
+
+
+if __name__ == "__main__":
+    main()

sectra_dicom_to_xml.py

@@ -0,0 +1,65 @@
+#!/usr/bin/env python3
+"""Convert a DICOM file to a simple XML representation.
+
+The script reads a `.dcm` file using ``pydicom`` and writes the contents to an
+XML file. Nested sequences are preserved so the resulting XML mirrors the
+structure of the DICOM dataset.
+
+Usage::
+
+    python sectra_dicom_to_xml.py -i path/to/file.dcm -o output.xml
+
+If no output file is given the name of the input file is used with the
+``.xml`` extension.
+"""
+
+import argparse
+import os
+import xml.etree.ElementTree as ET
+from xml.dom import minidom
+
+import pydicom
+from pydicom.dataset import Dataset
+
+
+def _add_dataset_to_element(dataset: Dataset, parent: ET.Element) -> None:
+    """Recursively append dataset contents as XML elements."""
+    for element in dataset:
+        tag = f"({element.tag.group:04X},{element.tag.element:04X})"
+        keyword = element.keyword or tag
+        attrs = {"tag": tag, "vr": element.VR, "name": keyword}
+        child = ET.SubElement(parent, "Element", attrs)
+        if element.VR == "SQ":
+            for item in element.value:
+                item_elem = ET.SubElement(child, "Item")
+                _add_dataset_to_element(item, item_elem)
+        else:
+            child.text = str(element.value)
+
+
+def dataset_to_xml(dataset: Dataset) -> ET.Element:
+    root = ET.Element("DicomDataset")
+    _add_dataset_to_element(dataset, root)
+    return root
+
+
+def dicom_to_xml(dicom_file: str, output_file: str | None = None) -> None:
+    ds = pydicom.dcmread(dicom_file)
+    root = dataset_to_xml(ds)
+    xml_str = minidom.parseString(ET.tostring(root)).toprettyxml(indent="  ")
+    if output_file is None:
+        output_file = os.path.splitext(dicom_file)[0] + ".xml"
+    with open(output_file, "w", encoding="utf-8") as f:
+        f.write(xml_str)
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="Convert DICOM to XML")
+    parser.add_argument("-i", "--input", required=True, help="Input DICOM file")
+    parser.add_argument("-o", "--output", help="Output XML file")
+    args = parser.parse_args()
+    dicom_to_xml(args.input, args.output)
+
+
+if __name__ == "__main__":
+    main()