How to get the number of series from a DICOM file? [closed] - dicom

Question

Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 4 years ago.
Improve this question
I have some 4-dimensional MR data in DICOM. The forth dimension can be time, b-value in DWI or whatever. How to determine how many slices and how many series in the forth dimension do I have?
For example, I have 400 images. How can I decide if there are 100 series and 4 slices or vice versa?
Edit:
I have figured it out by checking the slice position. If a given position repeats, I increment the number of stacks. My Python code below:
def getNumOfStacks(self, someImage):
sliceDict = dict()
for n in range(0, someImage.ImagesInAcquisition):
location = pydicom.dcmread(self.path+self.fileList[n]).SliceLocation
if location in sliceDict:
sliceDict[location] = sliceDict.get(location) + 1
else:
sliceDict[location] = 1
return list(sliceDict.values())[0]

Answer 1

The only way is to inspect the value of SeriesInstanceUID (tag number 0020,000e) for each single instance.
Depending on a tool you are using, the solution may be varied. For example, if you have dcmtk or gdcm, then in bash it would be like this:
find /path/to/dicom/files -exec dcmdump "{}" 2>/dev/null ";" | grep SeriesInstanceUID |sort -u
If you use gdcm, than put gdcmdump instead of dcmdump above.

Answer 2

MRI images in DICOM come in two different flavors:
"Traditional" MR Image Storage (SOP Class UID 1.2.840.10008.5.1.4.1.1.4)
Enhanced MR Image Storage (SOP Class UID 1.2.840.10008.5.1.4.1.1.4.1)
For both, like Bartlomiej wrote, the Series Instance UID can be used to determine which of the slices belong to the same series, and usually one series represents one stack of images.
For Enhanced MR, the concept of stacks was introduced. That is, a single DICOM object ("file") contains multiple frames ("images") which can be subdivided into stacks. In the Per Frame Functional Groups Sequence (5200,9230) you can find attributes which are specific for individual frames. In this case, you should read the Stack ID (0020,9056) and the In Stack Position Number (0020,9057) to seperate the stacks and order the slices within the stack.