I am currently an engineering student in Information Technology for Health. As part of a final internship study, I work with a company performing image processing on radiological images.
The principle is simple: an item of hospital PACS images in them to the company. The company changes the image pixels for simulations (and only the pixels of the image! This is very important for the future because I think no tag is changed at first). Then we send the image to hospitals. So there is an image that has exactly the same tags as the previous but different pixels.
Only after several tries, there are always problems during reintegration images to the hospital PACS. My internship is to allow the possible reintegration of these images in most cases.
Indeed, I saw that each hospital had its own PACS system and it would be probably impossible to find a way to reintegrate in all cases. However, after a little research month (about six of my internship) I am beginning to find some tracks and wanted to try to talk with people who know better than me.
I said that I take a placement test that took place two years ago. Therefore I only have very little information at the "error messages" or details on how the images are rejected by the PACS.
- I understood that each image was absolutely unique with a combination of 4 UID. I think these UIDs are a major cause for not inserting images, is it possible?
- Various tests were carried out with different hospitals summers and therefore with different PACS I guess. Do we know precisely the type of PACS that we want to integrate to address this issue?
- I also found a process to obtain a single IDU root to the company. Is absolutely necessary to follow this process because we are not creating medical images, we only change for simulations of radiation doses?
- I also found an explanation in the form of a UID, but I feel with several examples that no image UID really follows the same pattern. Why?
- If you have any other leads on what can block an image during insertion into a PACS, I'm interested.
Thank you very much in advance for your help.
General discussion for DICOM standard
1 post • Page 1 of 1