WIP feat(image-sets-normalization): add group by options

Also, remove import of "itkGDCMImageIO.h" by extracting CharStringToUTF8Converter

Author: PaulHax

packages/dicom/gdcm/CharStringToUTF8Converter.h

@@ -20,7 +20,6 @@
 
 #include <string>
 #include <set>
-#include "itkGDCMImageIO.h"
 
 using Tag = gdcm::Tag;
 using Tags = std::set<Tag>;

packages/dicom/gdcm/DICOMTagReader.h

@@ -0,0 +1,61 @@
+/*=========================================================================
+ *
+ *  Copyright NumFOCUS
+ *
+ *  Licensed under the Apache License, Version 2.0 (the "License");
+ *  you may not use this file except in compliance with the License.
+ *  You may obtain a copy of the License at
+ *
+ *         https://www.apache.org/licenses/LICENSE-2.0.txt
+ *
+ *  Unless required by applicable law or agreed to in writing, software
+ *  distributed under the License is distributed on an "AS IS" BASIS,
+ *  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ *  See the License for the specific language governing permissions and
+ *  limitations under the License.
+ *
+ *=========================================================================*/
+#ifndef TAGS_OPTION_PARSER_H
+#define TAGS_OPTION_PARSER_H
+
+#include "rapidjson/document.h"
+
+#include "Tags.h"
+
+std::optional<Tags> parseTags(itk::wasm::InputTextStream &tagsToRead, itk::wasm::Pipeline &pipeline)
+{
+    if (tagsToRead.GetPointer() == nullptr)
+    {
+        return std::nullopt;
+    }
+
+    rapidjson::Document inputTagsDocument;
+    const std::string inputTagsString((std::istreambuf_iterator<char>(tagsToRead.Get())),
+                                      std::istreambuf_iterator<char>());
+    if (inputTagsDocument.Parse(inputTagsString.c_str()).HasParseError())
+    {
+        CLI::Error err("Runtime error", "Could not parse input tags JSON.", 1);
+        pipeline.exit(err);
+        return std::nullopt;
+    }
+    if (!inputTagsDocument.HasMember("tags"))
+    {
+        CLI::Error err("Runtime error", "Input tags does not have expected \"tags\" member", 1);
+        pipeline.exit(err);
+        return std::nullopt;
+    }
+
+    const rapidjson::Value &inputTagsArray = inputTagsDocument["tags"];
+
+    Tags tags;
+    for (rapidjson::Value::ConstValueIterator itr = inputTagsArray.Begin(); itr != inputTagsArray.End(); ++itr)
+    {
+        const std::string tagString(itr->GetString());
+        Tag tag;
+        tag.ReadFromPipeSeparatedString(tagString.c_str());
+        tags.insert(tag);
+    }
+    return tags;
+}
+
+#endif // TAGS_OPTION_PARSER_H

packages/dicom/gdcm/Tags.h

@@ -43,15 +43,18 @@
 #include "itkMakeUniqueForOverwrite.h"
 
 #include "itkPipeline.h"
+#include "itkInputTextStream.h"
 #include "itkOutputTextStream.h"
 
-#include "DICOMTagReader.h"
+#include "CharStringToUTF8Converter.h"
 #include "Tags.h"
+#include "TagsOptionParser.h"
 #include "SortSpatially.h"
 
 std::string getLabelFromTag(const gdcm::Tag &tag, const gdcm::DataSet &dataSet)
 {
-  if (tag.IsPrivateCreator()) {
+  if (tag.IsPrivateCreator())
+  {
     return tag.PrintAsContinuousUpperCaseString();
   }
   std::string strowner;
@@ -529,7 +532,6 @@ using DicomFiles = std::unordered_set<DicomFile, dicomFileHash>;
 DicomFiles loadFiles(const std::vector<FileName> &fileNames)
 {
   DicomFiles dicomFiles;
-  itk::DICOMTagReader tagReader;
   for (const FileName &fileName : fileNames)
   {
     dicomFiles.insert(DicomFile(fileName));
@@ -559,20 +561,19 @@ bool compareTags(const gdcm::DataSet &tagsA, const gdcm::DataSet &tagsB, const T
   return true;
 }
 
-bool isSameVolume(const gdcm::DataSet &tagsA, const gdcm::DataSet &tagsB)
+bool isSameVolume(const gdcm::DataSet &tagsA, const gdcm::DataSet &tagsB, const Tags &criteria)
 {
-  const Tags criteria = {SERIES_UID, FRAME_OF_REFERENCE_UID};
   return compareTags(tagsA, tagsB, criteria);
 }
 
-Volumes groupByVolume(const DicomFiles &dicomFiles)
+Volumes groupByVolume(const DicomFiles &dicomFiles, const Tags &criteria = {SERIES_UID, FRAME_OF_REFERENCE_UID})
 {
   Volumes volumes;
   for (const DicomFile &dicomFile : dicomFiles)
   {
     const auto candidate = dicomFile.dataSet;
-    auto matchingVolume = std::find_if(volumes.begin(), volumes.end(), [&candidate](const Volume &volume)
-                                       { return isSameVolume(volume.begin()->dataSet, candidate); });
+    auto matchingVolume = std::find_if(volumes.begin(), volumes.end(), [&candidate, &criteria](const Volume &volume)
+                                       { return isSameVolume(volume.begin()->dataSet, candidate, criteria); });
 
     if (matchingVolume != volumes.end())
     {
@@ -587,16 +588,16 @@ Volumes groupByVolume(const DicomFiles &dicomFiles)
   return volumes;
 }
 
-ImageSets groupByImageSet(const Volumes &volumes)
+ImageSets groupByImageSet(const Volumes &volumes, const Tags &imageSetCriteria = {STUDY_UID})
 {
   ImageSets imageSets;
   for (const Volume &volume : volumes)
   {
     const gdcm::DataSet volumeDataSet = volume.begin()->dataSet;
-    auto matchingImageSet = std::find_if(imageSets.begin(), imageSets.end(), [&volumeDataSet](const Volumes &volumes)
+    auto matchingImageSet = std::find_if(imageSets.begin(), imageSets.end(), [&volumeDataSet, &imageSetCriteria](const Volumes &volumes)
                                          {
       const gdcm::DataSet imageSetDataSet = volumes.begin()->begin()->dataSet;
-      return compareTags(volumeDataSet, imageSetDataSet, {STUDY_UID}); });
+      return compareTags(volumeDataSet, imageSetDataSet, imageSetCriteria); });
     if (matchingImageSet != imageSets.end())
     {
       matchingImageSet->push_back(volume);
@@ -730,15 +731,25 @@ int main(int argc, char *argv[])
   std::vector<std::string> files;
   pipeline.add_option("--files", files, "DICOM files")->required()->check(CLI::ExistingFile)->type_size(1, -1)->type_name("INPUT_BINARY_FILE");
 
+  itk::wasm::InputTextStream seriesGroupByOption;
+  pipeline.add_option("--series-group-by", seriesGroupByOption, "Create series so that all instances in a series share these tags. Option is a JSON object with a \"tags\" array. Example tag: \"0008|103e\". If not provided, defaults to Series UID and Frame Of Reference UID tags.")->type_name("INPUT_JSON");
+  itk::wasm::InputTextStream imageSetGroupByOption;
+  pipeline.add_option("--image-set-group-by", imageSetGroupByOption, "Create image sets so that all series in a set share these tags. Option is a JSON object with a \"tags\" array. Example tag: \"0008|103e\". If not provided, defaults to Study UID tag.")->type_name("INPUT_JSON");
+
   itk::wasm::OutputTextStream imageSetsOutput;
   pipeline.add_option("image-sets", imageSetsOutput, "Image sets JSON")->required()->type_name("OUTPUT_JSON");
 
   ITK_WASM_PARSE(pipeline);
 
+  const std::optional<Tags> seriesGroupByParse = parseTags(seriesGroupByOption, pipeline);
+  const Tags seriesGroupBy = seriesGroupByParse.value_or(Tags{SERIES_UID, FRAME_OF_REFERENCE_UID});
+  const std::optional<Tags> imageSetGroupByParse = parseTags(imageSetGroupByOption, pipeline);
+  const Tags imageSetGroupBy = imageSetGroupByParse.value_or(Tags{STUDY_UID});
+
   const DicomFiles dicomFiles = loadFiles(files);
-  Volumes volumes = groupByVolume(dicomFiles);
+  Volumes volumes = groupByVolume(dicomFiles, seriesGroupBy);
   volumes = sortSpatially(volumes);
-  const ImageSets imageSets = groupByImageSet(volumes);
+  const ImageSets imageSets = groupByImageSet(volumes, imageSetGroupBy);
 
   rapidjson::Document imageSetsJson = toJson(imageSets);
   rapidjson::StringBuffer stringBuffer;

packages/dicom/gdcm/TagsOptionParser.h

@@ -18,12 +18,20 @@
 
 async def image_sets_normalization_async(
     files: List[os.PathLike] = [],
+    series_group_by: Optional[Any] = None,
+    image_set_group_by: Optional[Any] = None,
 ) -> Any:
     """Group DICOM files into image sets
 
     :param files: DICOM files
     :type  files: os.PathLike
 
+    :param series_group_by: Create series so that all instances in a series share these tags. Option is a JSON object with a "tags" array. Example tag: "0008|103e". If not provided, defaults to Series UID and Frame Of Reference UID tags.
+    :type  series_group_by: Any
+
+    :param image_set_group_by: Create image sets so that all series in a set share these tags. Option is a JSON object with a "tags" array. Example tag: "0008|103e". If not provided, defaults to Study UID tag.
+    :type  image_set_group_by: Any
+
     :return: Image sets JSON
     :rtype:  Any
     """
@@ -33,6 +41,10 @@ async def image_sets_normalization_async(
     kwargs = {}
     if files is not None:
         kwargs["files"] = to_js(BinaryFile(files))
+    if series_group_by is not None:
+        kwargs["seriesGroupBy"] = to_js(series_group_by)
+    if image_set_group_by is not None:
+        kwargs["imageSetGroupBy"] = to_js(image_set_group_by)
 
     outputs = await js_module.imageSetsNormalization(webWorker=web_worker, noCopy=True, **kwargs)
 

packages/dicom/gdcm/image-sets-normalization.cxx

@@ -18,12 +18,20 @@
 
 def image_sets_normalization(
     files: List[os.PathLike] = [],
+    series_group_by: Optional[Any] = None,
+    image_set_group_by: Optional[Any] = None,
 ) -> Any:
     """Group DICOM files into image sets
 
     :param files: DICOM files
     :type  files: os.PathLike
 
+    :param series_group_by: Create series so that all instances in a series share these tags. Option is a JSON object with a "tags" array. Example tag: "0008|103e". If not provided, defaults to Series UID and Frame Of Reference UID tags.
+    :type  series_group_by: Any
+
+    :param image_set_group_by: Create image sets so that all series in a set share these tags. Option is a JSON object with a "tags" array. Example tag: "0008|103e". If not provided, defaults to Study UID tag.
+    :type  image_set_group_by: Any
+
     :return: Image sets JSON
     :rtype:  Any
     """
@@ -55,6 +63,18 @@ def image_sets_normalization(
             pipeline_inputs.append(PipelineInput(InterfaceTypes.BinaryFile, BinaryFile(value)))
             args.append(input_file)
 
+    if series_group_by is not None:
+        pipeline_inputs.append(PipelineInput(InterfaceTypes.JsonCompatible, series_group_by))
+        args.append('--series-group-by')
+        args.append(str(input_count))
+        input_count += 1
+
+    if image_set_group_by is not None:
+        pipeline_inputs.append(PipelineInput(InterfaceTypes.JsonCompatible, image_set_group_by))
+        args.append('--image-set-group-by')
+        args.append(str(input_count))
+        input_count += 1
+
 
     outputs = _pipeline.run(args, pipeline_outputs, pipeline_inputs)
 

packages/dicom/python/itkwasm-dicom-emscripten/itkwasm_dicom_emscripten/image_sets_normalization_async.py

@@ -52,6 +52,7 @@ def test_ct():
     sorted_files = pick_files(image_sets[0])
     assert_equal(sorted_files, ct_series)
 
+
 def test_mr():
     assert mr_series[0].exists()
     out_of_order = [
@@ -67,6 +68,13 @@ def test_mr():
     assert_equal(sorted_files, mr_series)
 
 
+def test_series_group_by_option():
+    assert mr_series[0].exists()
+    group_by_tags = {"tags": ["0008|0018"]} # SOP Instance UID
+    image_sets = image_sets_normalization(mr_series, series_group_by=group_by_tags)
+    assert len(image_sets) == len(mr_series)
+
+
 def test_two_series():
     files = [
         orientation_series[1],

packages/dicom/python/itkwasm-dicom-wasi/itkwasm_dicom_wasi/image_sets_normalization.py

@@ -10,15 +10,23 @@
 
 def image_sets_normalization(
     files: List[os.PathLike] = [],
+    series_group_by: Optional[Any] = None,
+    image_set_group_by: Optional[Any] = None,
 ) -> Any:
     """Group DICOM files into image sets
 
     :param files: DICOM files
     :type  files: os.PathLike
 
+    :param series_group_by: Create series so that all instances in a series share these tags. Option is a JSON object with a "tags" array. Example tag: "0008|103e". If not provided, defaults to Series UID and Frame Of Reference UID tags.
+    :type  series_group_by: Any
+
+    :param image_set_group_by: Create image sets so that all series in a set share these tags. Option is a JSON object with a "tags" array. Example tag: "0008|103e". If not provided, defaults to Study UID tag.
+    :type  image_set_group_by: Any
+
     :return: Image sets JSON
     :rtype:  Any
     """
     func = environment_dispatch("itkwasm_dicom", "image_sets_normalization")
-    output = func(files=files)
+    output = func(files=files, series_group_by=series_group_by, image_set_group_by=image_set_group_by)
     return output

packages/dicom/python/itkwasm-dicom-wasi/tests/test_image_sets_normalization.py

@@ -10,15 +10,23 @@
 
 async def image_sets_normalization_async(
     files: List[os.PathLike] = [],
+    series_group_by: Optional[Any] = None,
+    image_set_group_by: Optional[Any] = None,
 ) -> Any:
     """Group DICOM files into image sets
 
     :param files: DICOM files
     :type  files: os.PathLike
 
+    :param series_group_by: Create series so that all instances in a series share these tags. Option is a JSON object with a "tags" array. Example tag: "0008|103e". If not provided, defaults to Series UID and Frame Of Reference UID tags.
+    :type  series_group_by: Any
+
+    :param image_set_group_by: Create image sets so that all series in a set share these tags. Option is a JSON object with a "tags" array. Example tag: "0008|103e". If not provided, defaults to Study UID tag.
+    :type  image_set_group_by: Any
+
     :return: Image sets JSON
     :rtype:  Any
     """
     func = environment_dispatch("itkwasm_dicom", "image_sets_normalization_async")
-    output = await func(files=files)
+    output = await func(files=files, series_group_by=series_group_by, image_set_group_by=image_set_group_by)
     return output

packages/dicom/python/itkwasm-dicom/itkwasm_dicom/image_sets_normalization.py

@@ -1,11 +1,17 @@
 // Generated file. To retain edits, remove this comment.
 
-import { BinaryFile } from 'itk-wasm'
+import { BinaryFile,JsonCompatible } from 'itk-wasm'
 
 interface ImageSetsNormalizationNodeOptions {
   /** DICOM files */
   files: string[] | File[] | BinaryFile[]
 
+  /** Create series so that all instances in a series share these tags. Option is a JSON object with a "tags" array. Example tag: "0008|103e". If not provided, defaults to Series UID and Frame Of Reference UID tags. */
+  seriesGroupBy?: JsonCompatible
+
+  /** Create image sets so that all series in a set share these tags. Option is a JSON object with a "tags" array. Example tag: "0008|103e". If not provided, defaults to Study UID tag. */
+  imageSetGroupBy?: JsonCompatible
+
 }
 
 export default ImageSetsNormalizationNodeOptions

packages/dicom/python/itkwasm-dicom/itkwasm_dicom/image_sets_normalization_async.py

@@ -53,6 +53,18 @@ async function imageSetsNormalizationNode(
       args.push(value as string)
     })
   }
+  if (options.seriesGroupBy) {
+    const inputCountString = inputs.length.toString()
+    inputs.push({ type: InterfaceTypes.JsonCompatible, data: options.seriesGroupBy as JsonCompatible })
+    args.push('--series-group-by', inputCountString)
+
+  }
+  if (options.imageSetGroupBy) {
+    const inputCountString = inputs.length.toString()
+    inputs.push({ type: InterfaceTypes.JsonCompatible, data: options.imageSetGroupBy as JsonCompatible })
+    args.push('--image-set-group-by', inputCountString)
+
+  }
 
   const pipelinePath = path.join(path.dirname(fileURLToPath(import.meta.url)), 'pipelines', 'image-sets-normalization')
 

packages/dicom/typescript/src/image-sets-normalization-node-options.ts

@@ -1,11 +1,17 @@
 // Generated file. To retain edits, remove this comment.
 
-import { BinaryFile, WorkerPoolFunctionOption } from 'itk-wasm'
+import { BinaryFile,JsonCompatible, WorkerPoolFunctionOption } from 'itk-wasm'
 
 interface ImageSetsNormalizationOptions extends WorkerPoolFunctionOption {
   /** DICOM files */
   files: string[] | File[] | BinaryFile[]
 
+  /** Create series so that all instances in a series share these tags. Option is a JSON object with a "tags" array. Example tag: "0008|103e". If not provided, defaults to Series UID and Frame Of Reference UID tags. */
+  seriesGroupBy?: JsonCompatible
+
+  /** Create image sets so that all series in a set share these tags. Option is a JSON object with a "tags" array. Example tag: "0008|103e". If not provided, defaults to Study UID tag. */
+  imageSetGroupBy?: JsonCompatible
+
 }
 
 export default ImageSetsNormalizationOptions

packages/dicom/typescript/src/image-sets-normalization-node.ts

@@ -60,6 +60,18 @@ async function imageSetsNormalization(
       args.push(name)
     }))
   }
+  if (options.seriesGroupBy) {
+    const inputCountString = inputs.length.toString()
+    inputs.push({ type: InterfaceTypes.JsonCompatible, data: options.seriesGroupBy as JsonCompatible })
+    args.push('--series-group-by', inputCountString)
+
+  }
+  if (options.imageSetGroupBy) {
+    const inputCountString = inputs.length.toString()
+    inputs.push({ type: InterfaceTypes.JsonCompatible, data: options.imageSetGroupBy as JsonCompatible })
+    args.push('--image-set-group-by', inputCountString)
+
+  }
 
   const pipelinePath = 'image-sets-normalization'