how to determine the pre-contrast and post-contrast through series time, acquisition time and content time in DCE-MRI dicom images

Support
1

I was studying the DCE-MRI image on Multi-center breast DCE-MRI data and segmentations from patients in the I-SPY 1/ACRIN 6657 trials (ISPY1) - The Cancer Imaging Archive (TCIA) Public Access - Cancer Imaging Archive Wiki

It seems that they have pre-contrast, early post-contrast, and late post-contrast images, all of which are stored in the same DICOM series or multiple folders. However, I cannot determine which image is pre-contrast and which ones are post-contrast. In the DICOM tags, I found four time-related tags: study time, series time, acquisition time, and content time. I assume that the series time indicates the time when the pre-contrast image was collected. So, how can I determine which series belongs to pre-contrast and which series belongs to post-contrast using these time tags?

2

Determining this info is not always straightforward, and for older DICOM data (the data in the ISPY1 study is from 1980s) it can be non-standard. The TCIA does a great job of including supporting documentation. The link you sent has a link to the Study Protocol and it states that
“multiple quantitative parameters are extracted for estimation of
tumor volume and parameters related to contrast kinetics: initial percent
enhancement, PE = (S1-S0)/S0, and signal enhancement ratio (SER)=(S1-S0)/(S2-
S0), where S0, S1 and S2 are the pre-contrast, first post-contrast and second-post
contrast signal intensities of the three time-point data acquisition.27”

Looking through the dicom data, for each patient there is DICOM series labeled
Dynamic-3dfgre
It contains all three time points, in sequence. It has 180 time-sequential slices representing 3 passes over the breast. These are the S0, S1, and S2 images described in the study. You can view the data using this link.

There are then two secondary/derived series:
Dynamic-3dfgre:SER
and
Dynamic-3dfgre: PE1
These were computed as defined in the study protocol.

I haven’t verified, but the acquisition time for each of the DICOM objects (slices) in the Dynamic-3dfgre series should allow you to differentiate S0 from S1 from S2.

Hope that helps.

3

Great. Though some error exists, for instance, 095917 and 095918 for the acquisition time of one pre-contrast series. But Seems like the acquisition time is the only information to differentiate this. Cannot find other related tags.Thank you so much for your help.