Datastore-Local

Copyright (c) General Electric Company, 2017. All rights reserved.

Local storage implementation of a Rt 106 datastore. This is a simple datastore using a directory structure on the filesystem to manage data. This datastore is geared for small, simple deployments. It provides a simple mechanism to wrap or ingest pre-existing data into Rt 106. The rt106-datastore REST API is used to serve the data to applications and algorithms.

Local filesystem organization

rt106-datastore-local prescribes a folder structure for data to manage primary (source) data as well as derived data from algorithm executions.

• Patients
  • Patient ID (or Patient Name)
    • Primary
      • Imaging
        • Study ID
          • Series ID
            • Image
            • Image
          • ...
        • ...
      • Tables (future)
      • Monitoring (future)
      • Records (future)
    • Results
      • Pipeline ID
        • Execution ID
          • Imaging
            • Study ID
              • Series ID
                • Image
                • Image
              • ...
            • ...
          • Tables (future)
          • Monitoring (future)
          • Records (future)
• Slides
  • Slide
    • Region
      • Source
        • Channel
          • Image
      • Pipeline ID
        • Execution ID
          • Channel
            • Image
      • Pipeline ID
        • Execution ID
          • Channel
            • Image

ID's may be DICOM UIDs (Study Instance UID) or can be UUIDs.

Docker container

To build the docker container for the front-end:

$ docker build -t rt106/rt106-datastore-local:latest .

If you use HTTP proxies in your environment, you may need to build using

$ docker build -t rt106/rt106-datastore-local:latest  --build-arg http_proxy=$http_proxy --build-arg https_proxy=$https_proxy  --build-arg no_proxy=$no_proxy .

Example radiology data

A sample of the data from the Visible Human Project can be downloaded and arranged as a local datastore by passing the environment variable DOWNLOAD_RAD_DEMO_DATA='on' or DOWNLOAD_RAD_DEMO_DATA='force' to the rt106-datastore-local container on startup.

(DICOM files for Visible Human Project data are downloaded from https://mri.radiology.uiowa.edu/VHDicom.)