View volumetric (3D) medical images in Jupyter notebooks

This tiny, but very useful utility enables interactive slice-by-slice viewing of 3D images in ipython notebooks. It builds upon ImageSliceViewer3D and is specifically designed for viewing CT images. Key features:

• Change orientation of 3D images to view them along any of the three possible axes
• Window CT images. All major presets for window level and window width provided. See the table below for the complete list of presets
• Window CT images using custom values for window level and window width

CTViewer can be installed from PyPI with the following command:

  pip install ctviewer

CTViewer takes as input a 3-dimensional numpy array. For windowing to work as intended, the voxel values should be in Hounsfield Units (HU). Typical usage:

  from ctviewer import CTViewer
  # assuming that volumetric_image is the 3-dimensional numpy array
  CTViewer(volumetric_image)

For a more detailed example of loading a stack of dicom images from disk, converting to HU, and then viewing using CTViewer, check the sample_run.ipynb inside the examples folder

The following windows are available as presets, along with custom inputs for window level and window width:

Reach out to me at one of the following places!

Twitter: @vibhuagrawal
Email: vibhu[dot]agrawal14[at]gmail