Medical Shape

A torchio extension for shapes and their processing.

Usage

medical_shape provides 3 major classes for usage:

Shape

Shape is a subclass from torchio.data.Image. It stores arbitrary pointclouds together with their descriptions and an associated affine matrix. The pointclouds are stored as 2D torch.Tensor in the form NxD where N is the number of points and D is the dimensionality of points (usually D=3).

ShapeSupportSubject

The ShapeSupportSubject is an extension of torchio.data.Subject to allow the inclusion of Shape-type objects into the subject. It should be used instead of torchio.data.Subject, whenever a shape is included and is also safe to use without a shape (will behave exactly like torchio.data.Subject in that case) as shapes often require special handling.

TransformShapeValidationMixin

This class is a transformation mixin to allow checks whether shape-support is required and to raise warnings if the incorrect Subject-type was used. All transformations supporting shapes should inherit from it (and as it is derived from torchio.transforms.Transform, it is also safe to use this class as a standalone baseclass).

Installation

This project can be installed either from PyPI or by cloning the repository from GitHub.

For an install of published packages, use the command

    pip install medical-shape

To install from the (cloned) repository, use the command

    pip install PATH/TO/medical-shape

You can also add -e to the command to make an editable install in case you want to modify the code.

You can also install the package directly from GitHub by running

    pip install git+https://github.com/justusschock/medical-shape.git

Docker Images

We provide a docker image for easy usage of the package and as a base image for other projects.

The file for this image can be found at dockers/Dockerfile. We provide both, a CPU-only and a CUDA-enabled image based on the NVIDIA NGC PyTorch image. These images can be found on DockerHub.