I am new to Dicom and Orthanc. We have a Zeiss HFA 745 in our practice and I am looking for a way to transfer the data to an Orthanc server. I have configured the interface on the device and on the server, but I get the following error message in Orthanc:
These kind of errors typically appear when something tries to access Orthanc on port 4242 with another protocol than DICOM (like HTTP) or maybe with DICOM TLS enabled instead of plain DICOM.
Any DICOM compatible device is supposed to work with Orthanc so you should also feel free to contact your Zeiss support - your probably pay them for that .
Worst case scenario, record the TCP traffic on the DICOM Port with tcpdump or wireshark and share it here but make sure not to share any sensitive information …
They are available in the mainline binaries and will be part of the next official release.
Hope this helps,
Alain.
Free software isn’t free to make — it takes time, effort, and resources. If you appreciate the project, consider donating to help sustain its development!
I checked them, but unfortunately they do not seem to solve our current issue.
The main problem is apparently not standard DICOM query/retrieve behavior inside Orthanc. The HFA II-i 745 seems to use a proprietary ZEISS protocol for export:
for “report / examination” export, it sends proprietary TCP/XML
for EMR/PMS export, it writes XML/PDF over SMB and expects a proprietary response/handshake
Our current workaround is therefore:
receive the proprietary export
extract XML / raw data / PDF
convert it into proper OPV DICOM
send that to Orthanc
This already works partially: we can receive the payload and import the resulting data into Orthanc.
The remaining blocker is that the HFA device still does not accept our server/gateway responses as valid, so the device reports incomplete or failed transfer even though the data can already be captured on the server side.
So my impression is that your new C-FIND fixes are probably relevant for later Zeiss query/retrieve or MWL integration, but they do not address the current export acknowledgment problem we are stuck on.