What's the difference between Instance Number and Acquisition Number tags in DICOM images?

I need explanations about those two tags and what values are appropriate to insert if they are not given

Answer 1

[Rik]

This is not a Matlab question, but a question about DICOM standards. You can easily find the descriptions by using your favorite internet search engine.

link to NEMA standard

Attribute name: Acquisition Number
Attribute tag: (0020,0012)
Attribute type: 3 (Optional Data Element)
Description:
A number identifying the single continuous gathering of 
data over a period of time that resulted in this image.

Attribute name: Instance Number
Attribute tag: (0020,0013)
Attribute type: 2 (Required Data Element)
Description:
A number that identifies this image.
Note
This Attribute was named Image Number in earlier versions of this Standard.

So the acquisition number can be used to identify slices/single images within a series, while the instance number identifies the complete image.

[Rik]

Did this answer your question, or do you have a follow-up question about these tags?

Answer 2

[Paul Wright]

Just because this ranked high in a search, and I've been working on this, I wanted to chip in. There may be several Instance Numbers for one Acquisition Number. For example, in axial CT data, the acquisition refers to the block of slices from one cycle of the detector, whereas the instance number identifies each slice within that block. The "complete image" in this case would be the CT volume, made up of multiple acquisitions.

Note that instance number does not have to be either sequential or unique, so you can't rely on it to sort slices, but you can look at Image Position Patient or SOP Instance UID to get slice position or unique slice ID. Images that are part of the same 3D volume share the same Series Instance UID.