If done correctly you should see all the containers running as belowĤ. As per instructions build the three containers and configure with strong password using docker compose.Read the instructions here and pull the containers.Install Docker Desktop (stable release for windows).The notebook I used for the python commands can be found here The instructions in GitHub were quite limited for the docker version. I decided to try test it out (and write down the steps for anyone else that wanted to follow along). It allows rapid access to images using restful services (QIDO-RS, STOW-RS etc) which allows for pipeline creation.Good for the initial ingestion of images either from the PACS or directly from imaging modalities. It works great with older PACS systems who only support dicom composite commands commands (C-Find, C-Store etc).I have used an instance of Dcm4chee in production for a similar purpose ( anomaly detection) as it supports both dicom composite commands and dicom web communication meaning Traditionally been an issue with legacy PACS systems which rely on dicom composite commands for communication (rather than restful services) which silos the information and limits the functionality that can be performed.Īlthough this medical imaging archive can easily be spun up on Microsoft Azure you can also access the docker version which allows it to be locally hosted and run. This architecture is suppose to help unlock access to images/data, most likely for machine learning classification models.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |