1 -----------------------------------------------------------------------------
2 Proposed Template of an entry in this TODO:
3 (Note: Date is the date of registering of first demand.)
13 -----------------------------------------------------------------------------
14 Description: GDCM does not read Spacing properly.
18 GDCM is ready completely random values to find out the spacing/origin or an
19 image. It *SHOULD* first find the IOD and then deduce which tags to read
20 Eg: Cross section this is in Pixel Spacing (0028,0030)
21 CR is in Imager Pixel Spacing (0018,1164)
22 US is in Pixel Ratio (0028,0034)
24 (3002,0011) Image Plane Pixel Spacing
25 (3002,0012) RT Imaga ePosition
27 (3004,000c) for deducing Z spacing
28 -----------------------------------------------------------------------------
29 Description: GDCM is pretty dumb with binary data
33 D 0028|0009 [AT] [Frame Increment Pointer] [gdcm::Binary data loaded;length = 4]
34 -----------------------------------------------------------------------------
35 Description: gdcm does not handle deflate TS
39 image_dfl from dclunie website is not working since the *whole* dataset
40 is compressed. Therefore we cannot even parse the file.
41 -----------------------------------------------------------------------------
42 Description: Problem with PrintFile and OB field
46 Doing a PrintFile on RickAvila DataSet we get:
47 D 0043|1028 lg : x(2) 2 Off.: x(bb2) 2994 [OB] [Unique image iden] [00]
48 Whereas hexedit shows 3030 ??
49 -----------------------------------------------------------------------------
50 Description: Make sure that that gdcm lib can be dlopen'ed/dlclose
54 As reported in ITK mailing list, the global symbols are completely
55 skrewed up in gdcm, preventing one to dlopen/dlclose gdcm lib.
56 -> Need to add a nighlty test
57 -> Need to write the code :(
58 -----------------------------------------------------------------------------
59 Description: Add testing of valid dictionary
63 Since that now private/shadow dictionary are available
64 it would be nice to start implementing an automatic mode of loading
65 those dictionary as we read the public one.
66 Comments: It shouldn't be too difficult to 'manualy' ask memory merging
67 of a Private Dict into the public one (I can do it soon).
68 Automatic recognition of the Private Dict to be used seems hopeless.
69 -----------------------------------------------------------------------------
70 Description: gdcm::SerieHelper / UID / set of rules
74 gdcm::SerieHelper now offer a mechanism to add rules to sub select image
75 as we iterate over them within a subdirectory structure. But what if UID was too
76 restrictive ? Therefore UID subselection become only a good default rule, which
77 should ideally be removed when user need a specific task.
79 -----------------------------------------------------------------------------
80 Description: Add testing of valid dictionary
84 It is potentially possible that user modifies the dictionary that gdcm provides
85 and this is also possible that the dictionary generated from pdf is buggy
86 (see 2001,xx5F. VR = SQ, VM = 1-n, from
87 www.medical.philips.com/main/company/connectivity/assets/docs/dicomcs/mr91.pdf)
88 Therefore gdcm should check for any typo, and report it (if possible)
90 -----------------------------------------------------------------------------
91 Description: Add kwsys as a subdir somewhere in gdcm
95 kwsys is a lightweight library developped by kitware, used in project like
96 ITK, VTK, CMake and ParaView. It runs and compile on almost any plateform with c++
97 compiler. And it provide a cross plateform approach to any kind of system call
98 (executing a process, killing a process, realpath, filename/directory management ...)
100 * jpr : what does 'kwsys' stand for?
101 -----------------------------------------------------------------------------
102 Description: Extent reading support
106 Comments: All ITK/VTK readers support selecting extent. gdcm should support
107 selecting extent before being inserted into ITK
108 -----------------------------------------------------------------------------
109 Description: gdcmDicomDir and SQItem creation
112 Details: DicomDir creates some SQItem (by new). Then, it creates
113 the corresponding DicomDirPatient, etc. using the content of the SQItem
114 (the content is composed with some DocEntry's that can't be destroyed).
115 So, if the SQItem is deleted, then it's content is deleted to. But the
116 DicomDirPatient, etc. use the content of the SQItem. Then, the SQItem can't
117 be deleted, and when have memory leaks
120 -----------------------------------------------------------------------------
121 Description: [BUG] Better handling of unfound Dicom dictionary.
122 When gdcm doesn't find the Dicom dictionary (because it's
123 path to the directory of dictionary is uncorrect, either
124 because the install relative layout was broken after file moves
125 or because the environnement variable GDCM_DICT_PATH is
126 unpropely set), gdcm will:
128 2/ throw an exception (that is internaly UNcatched by gdcm)
129 that in most cases provoques the caller application to
130 exit uncleanly (uncatched excpetions in fine call abort() ).
131 Additionaly on Win32 the warning print isn't displayed because
132 exiting occurs prior to cerr or cout is flushed properly.
135 Details: fixes (from dirty to clean)
136 1/ force Win32 to flush it's buffer so at least the user gets some
137 reason why it's application exited (when called in command
138 environement). Note: it looks like the "cerr << flush" fails. Sigh.
139 2/ within gdcm catch the exception, display a decent warning, and
141 3/ see the comment below on how to enhance the API and fix things
143 Comments: ENH proposal:
144 The caller migth not be aware of the path to the dictionaries
145 on invocation of gdcm (think this path is set by the Interface
146 because the caller wants to skip the default gdcm dictionary in order
147 to provide his own ones e.g. another language based one).
148 Hence, gdcm should postpone the parsing of the default dictionary
149 instead of doing it on library entry.
150 This would enable two things:
151 - It would give a chance to the caller to set the path to
152 the dictionaries he whishes to use, through a call to
153 newly created DictSet::SetDictionaryPath( string ).
154 - It would avoid the burden of using the GDCM_DICT_PATH
155 environnement variable and enable GDCM CONTROL FROM WITHIN
156 THE API. Optionaly, if the caller didn't use the API to
157 provide his prefered path, gdcm could still default to
159 -----------------------------------------------------------------------------
160 Description: vtk/vtkGdcmReader doesn't positionate the normal to the image
163 Details: vtkGdcmReader::CheckFileCoherence() sets the DataOrigin[i]
164 but doesn't set the plane (image seen in 3D) normal (is it
165 possible any how). This plane normal could be extracted from
166 the "orientation" info of the gdcmHeader ( refer to
167 grep "Orientation" Dicts/dicomV3.dic).
168 Problem exhibiting this defect: cine loop on a stack of images
169 whose Origin is correct, but whose normal is not set will
172 * vtkGdcmReader inherits from vtkImageReader which aggregates
173 a vtkTranform. vtkGdcmReader could store (when the user requires
174 it, see below) the origin/normal taken from the Dicom Header
175 within this vtkTransform (looks like a natural place to store
176 this spacial information).
177 * Both settings of the origin and/OR the normal of the plane (image)
178 should be an option defined with a flag (On/Off) in the
180 -----------------------------------------------------------------------------
181 Description: test the private dictionary part.
186 * Frog: where can we obtain such a private/dictionary and the corresponding
187 Dicom file ? Any examples on-line ?
188 * jpr : some are in gdcm/Dicts (built from pdf documents found on constructors'
190 When we check them against existing images, we see the are uncomplete
192 -----------------------------------------------------------------------------
193 Description: generate methods based on VM.
197 * cmake should parse le DICOM dictionary to generate methods like
198 gdcm???::SetImagePosition(int, int)
200 //generated content do not edit
203 gdcm???::SetImageNumber(int)
205 //generated content do not edit
209 * Regrain: a dicom dictionary entry name is NOT UNIQUE [this means
210 two tags=(group, element) can share the same name].
211 What should the wrapper do in such a case !?
212 * Frog: what does VM stand for ?
213 * VM = Value Multiplicity
214 -----------------------------------------------------------------------------
215 Description: Add information on supported imagers (constructor/model)
218 Details: in order to promote gdcm make a list (on the web pages)
219 of images successfully parsed based on a constructor/model ordering
221 * frog: gdcmData only lists pathological images. How to collect
222 the ones gdcm works smoothly with (hopefully gdcmData is a small
223 subset of what we would like).
224 * jpr : gdcmData contains images that caus*ed* us some troubles.
225 the aim of gdcm is to read *all* the images, from *all* the
226 constructors and *all* the models.
227 Better we do a 'gdcm Dicom Hall of Shame' with bugged header images,
228 explaining *why* the header is bugged.
229 -----------------------------------------------------------------------------
230 Description: Add a GetVersion() global function.
233 Details: This is to be used for version assertion with gdcmPython
234 Comments: Done (August 2005)
235 -----------------------------------------------------------------------------
236 * vtk/vtkGdcmHeader.cxx: if speed becomes a concern some changes can
237 be made at the cost of memory consumption (refer to header of
238 vtk/vtkGdcmHeader.cxx)
239 -----------------------------------------------------------------------------
240 * gdcmElValSet::SetElValueLengthByNumber IMNSHO should be trashed.
241 It's only purpose is a onliner substitute to calling GetElValueByNumber
242 and then SetLength. This only obfuscates the caller code more than
244 Besides the definition of gdcmElValSet::SetElValueLengthByNumber itself
245 it quite poor since it is a almost exact copy of
246 gdcmElValSet::GetElValueByNumber except for the returned code.
247 gdcmHeader::SetPubElValLengthByNumber (which is based on
248 gdcmElValSet::SetElValueLengthByNumber) is used nowhere...
250 * jpr : all the methods SetxxxByName were trashed.
251 all the methods SetxxxByNumber were renamed
252 A general method clean out was performed
253 -----------------------------------------------------------------------------
254 * All (or at least many of) the methods of gdcmHeader whose only arguments
255 are an ElValue* (e.g. FindLength, FindVR, LoadElementValue...) can
256 be moved away to ElValue class on condition of transmitting the
257 gdcmHeader.fp attribute. This change should be considered since it
258 would allow those method to avoid artificial calls to ElValue::GetElement(),
260 -----------------------------------------------------------------------------
261 * Group length is not a unique tag in a file. Hence avoid putting it
262 in the element values dictionary without doing something smarter
263 (say, instead of storing the length store the group and the length
264 so we can related a length to a group).
265 -----------------------------------------------------------------------------
266 * GetPubElValByNumber doit faire la difference entre chaine vide
267 et chaine pas trouve'e. Eventuellement raiser une exception ?
268 -----------------------------------------------------------------------------
269 * gdcmHeader::LoadElements only loads the element whose length is
270 below the specified size. When accessing the value of such an element
271 the content is unfound ! Find a decent way of loading the value on
273 -----------------------------------------------------------------------------
274 * JPR: supply a method that only reads/loads (?) the Dicom elements
276 -----------------------------------------------------------------------------
277 * JPR: gdcmHeader::CheckSwap() dans le cas ACR pas propre, degager tout de
278 suite si on a deduit que c'en est pas...
279 -----------------------------------------------------------------------------
280 * python /usr/lib/python2.2/site-packages/DaVaW/demo/dvwDcmReader.py
281 and load image /home/frog/cvs/DCMlib/Data/CT-MONO2-16-ankle.dcm
282 will yield wrong coloring scheme as opposed to
283 affim filein=/home/frog/cvs/DCMlib/Data/CT-MONO2-16-ankle.dcm
284 -----------------------------------------------------------------------------
285 * gdcmFile should implement the following API:
287 WriteDicom.SetFileName("MyDicomFile.dcm");
288 string * AllTags = gdcmHeader.GetDcmTagNames();
289 WriteDicom.SetDcmTag(AllTags[5], "253");
290 WriteDicom.SetDcmTag("Patient Name", "bozo");
291 WriteDicom.SetDcmTag("Patient Name", "bozo");
292 WriteDicom.SetImageData(Image);
295 Anonymize(ostream& output) {
296 a = gdcmFile("toto1");
297 a.SetPubValueByName("Patient Name", "");
298 a.SetPubValueByName("Date", "");
299 a.SetPubValueByName("Study Date", "");
302 -----------------------------------------------------------------------------