medviz is a library and a program for extracting 2D images (or
"slices") from raw 3D volumetric data.
medviz has a simple to use command-line interface. Run cargo run -- --help or medviz --help after building and installing to get help.
Produce BMP image files using very verbose logging and short option
names: medviz -vvv -m tests/data/sinus.mhd -d tests/data/sinus.raw -z z.bmp -y y.bmp -x x.bmp
Produce RAW files using very verbose logging and long option names:
medviz -vvv --metadata tests/data/sinus.mhd --data tests/data/sinus.raw --zfile z.raw --yfile y.raw --xfile x.raw --raw
Cargo can be used to install medviz into ~/.cargo/bin: cargo install --path .
To test the medviz library, you need to have the test files
available which are, by default, compressed using zpaq in an
lrz-tar archive and provided under tests/data.tar.lrz.
The reason for this is that Github only allows checked-in files up to 50MB. I previously used Git LFS to store those files but immediately exhausted the provided large-file quota.
To extract the test files, cd into the tests directory and run
lrzuntar data.tar.lrz, you should end up with a tests/data
directory containing .bmp and .raw files.
Once the test data is available, execute cargo test in the top-level
source directory.
The test files are produced by a normal run of medviz and were
manually checked. The integration test checks the current execution
against these "model" files. I don't currently know a better way to
system test a tool like medviz.