Skip to content

Setup a multi-arch Diana DICOM-handler service on embedded systems

Notifications You must be signed in to change notification settings

derekmerck/reDiana

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 

Repository files navigation

Remote Embedded DIANA Setup

Derek Merck
derek_merck@brown.edu
Rhode Island Hospital and Brown University
Providence, RI

Setup a multi-arch DIANA DICOM-handler service on embedded systems.

Use It

$ docker-compose up

When using Resin.io devices in "local mode" for development and testing, you can also direct the device's Docker daemon remotely.

$ DOCKER_API_VERSION=1.22 DOCKER_HOST=tcp://device_ipl:2375 docker-compose up -d

Docker Image Dependencies

Keras-TF is used instead of Conda for machine learning configurations.

Supported Architectures

Tested on amd64 virtual machines and arm32v7 Raspberry Pi. Can be compiled for aarch64, but currently untested.

Dependencies are are all multi-architecture, so the image name alone is typically sufficient to find the correct image when pushing to a range of devices. When uploading via Resin.io for Raspberry Pi, though, you need to add explicit "arm32v7" tags. Their builder is aarch64 and will look for an "arm64v8" tag by preference, rather than the base image target architecture, as it should.

License

MIT

About

Setup a multi-arch Diana DICOM-handler service on embedded systems

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages