A Radiology Service for Processing Pipelines
The RADS2 team has provided support to Radiology through creating a pipeline for the delivery of EHR data. The data made available by the RADS2 team is used by the Niffler framework. This data is retrieved from the Cerner database and stored locally in the RTA Oracle database. The Radiology group periodically (set at every 30 minutes) acquires labs, orders, and medications from the local RTA database through REST APIs developed by the RADS2 team.
Niffler is a lightweight framework to facilitate executing machine learning pipelines and processing workflows on DICOM images and metadata. Niffler facilitates efficient transfer of DICOM images on-demand and real-time from PACS to the research environments. Niffler is also integrated with the radiology information system (RIS) to get clinical data in real-time. The DICOM images from the PACS and clinical data retrieved from the RIS can be used in conjunction in real-time as well as retrospectively on-demand.
The RTA team has implemented the middleware deployment depicted in the figure, together with the Emory Radiology.
- Clinical data (labs, orders, and medications) acquired from the RIS via a database link.
- Integration with the PACS data at the research cluster.
- The data is deleted nightly at the research cluster after the processing workflows are executed.
Several pilot studies have been ongoing and completed, including an IVC filter detection and scanner utilization computation, as part of the Niffler workflows using PACS and RIS data.
Design and Implementation
