glenoidplanefitting

Authors: Asta Olafsdottir, Stephen Thompson

glenoidplanefitting provides tools for measuring the Glenoid version, useful when planning reconstructive shoulder surgery. glenoidplanefitting is part of the scikit-surgery software project, developed at the Wellcome EPSRC Centre for Interventional and Surgical Sciences, part of University College London (UCL), in collaboration with surgeons at the Royal National Orthopaedic Hospital

glenoidplanefitting is tested on Python 3.x, but may work with other versions of Python

glenoidplanefitting currently requires the user to manually identify relevant anatomical landmarks on the scapula and glenoid dish. These points can then be passed to glenoidplanefitting to calculate the Glenoid version using multiple methods. Example usage:

python glenoidplanefitting.py --fried_points landmark_friedman.fcsv --output friedman.vtp --visualise glenoid.vtp

glenoidplanefitting is a work in progress with plans to automate the point picking to create software to enable a large scale comparison of different methods of Glenoid version methods on a clinical data set. If you are interested in contributing to this work please get in touch or follow the guides below.

Developing

Cloning

You can clone the repository using the following command:

git clone https://github.com/SciKit-Surgery/glenoidplanefitting

Running tests

We use tox to run tests and run static code analysis using Lint.

pip install tox
python -m tox

Installing

You can pip install directly from the repository as follows:

pip install git+https://github.com/SciKit-Surgery/glenoidplanefitting

Contributing

Please see the contributing guidelines.

Useful links

Licensing and copyright

Acknowledgements

Supported by Wellcome and EPSRC.

Dependency Graph

digraph prof {
ratio = fill;
node [style=filled];
"SciKit-SurgeryGlenoid" -> "NumPy";
"SciKit-SurgeryGlenoid" -> "SciKit-SurgeryVTK" [color="0.515 0.762 0.762"];
"SciKit-SurgeryGlenoid" -> "SciKit-SurgeryCore" [color="0.515 0.762 0.762"];
"SciKit-SurgeryGlenoid" -> "VTK";

"SciKit-SurgeryVTK" -> "NumPy";
"SciKit-SurgeryVTK" -> "VTK";
"SciKit-SurgeryVTK" -> "PySide2";
"SciKit-SurgeryVTK" -> "OpenCV-contrib-python";
"SciKit-SurgeryVTK" -> "SciKit-SurgeryCore" [color="0.515 0.762 0.762"];
"SciKit-SurgeryVTK" -> "SciKit-SurgeryImage" [color="0.515 0.762 0.762"];

"SciKit-SurgeryImage" -> "NumPy";
"SciKit-SurgeryImage" -> "OpenCV-contrib-python";
"SciKit-SurgeryImage" -> "SciKit-SurgeryCore" [color="0.515 0.762 0.762"];
"SciKit-SurgeryCore" -> "NumPy";

"SciKit-SurgeryGlenoid" [color="0.515 0.762 0.762"];
"SciKit-SurgeryVTK"[color="0.515 0.762 0.762"];
"SciKit-SurgeryImage"[color="0.515 0.762 0.762"];
"SciKit-SurgeryCore"[color="0.515 0.762 0.762"];
} — glenoidplanefitting’s Dependencies

latest

glenoidplanefitting package

Subpackages

glenoidplanefitting.algorithms package

Submodules

glenoidplanefitting.algorithms.colour_palette module

A colour palette based on Bang’s colour pallette Wong, B. Points of view: Color blindness. Nat Methods 8, 441 (2011). https://doi.org/10.1038/nmeth.1618

glenoidplanefitting.algorithms.colour_palette.bang_list()[source]: Returns the bang palette as a list

glenoidplanefitting.algorithms.friedman module

This is an implentation of Friedman’s method, see

Friedman RJ, Hawthorne KB and Genez BM. The use of computerized tomography in the measurement of glenoid version. J Bone and Joint Surg Am 1992; 74: 1032–7.

glenoidplanefitting.algorithms.friedman.create_friedman_line(point0, point1)[source]

Determines the second point needed to form the Friedman line

Parameters

point0 – First point on glenoid line, anatomically defined as a point on the anterior margin of glenoid
point1 – Second point on glenoid line anatomically defined as a point on the posterior margin of glenoid

Raises

Value Error if the z values of point0 and point1 are not equal

Returns

The midpoint of the glenoid line, which is the second point of the Friedman line

glenoidplanefitting.algorithms.friedman.friedman_version(glenoid0, friedman1, friedman0)[source]

Determines the glenoid version using the Friedman line

Parameters

glenoid0 – First point on glenoid line, anatomically defined as a point on the anterior margin of the glenoid
friedman1 – Second point on the Friedman line, anatomically defined as the midpoint of the glenoid fossa
friedman0 – First point on the Friedman line, anatomically defined as the medial tip of the scapula

Returns

The glenoid version (positive value indicates retroversion)

glenoidplanefitting.algorithms.models module

Functions to create vtk models for visualisation of results

glenoidplanefitting.algorithms.models.make_friedman_model(point1, point2)[source]

Makes a vtk line source from two set points

Parameters

point1 – one end of the line
point2 – other end of the line

Returns

The line

glenoidplanefitting.algorithms.models.make_plane_model(plane_centre, normal_vector, resolution=10, plane_size=200.0)[source]

Makes a vtk plane source, with centre and normal vector

Parameters

plane_centre – a point on the plane
normal_vector – the plane normal vector

Returns

The plane

glenoidplanefitting.algorithms.models.make_sphere_model(point, radius=5.0)[source]

Make a sphere source which we can use to represent a landmark point

Parameters: point – the point

:returns the vtkPointSource

glenoidplanefitting.algorithms.models.make_vault_model(point1, point2)[source]

Makes a vtk line source from two set points

Parameters

point1 – one end of the line
point2 – other end of the line

Returns

The line

glenoidplanefitting.algorithms.plane_fitting module

This is an implementation of a two plane method, see

A. Ganapathi, J. McCarron, Chen, J. Iannotti. Predicting normal glenoid version from the pathologic scapula: a comparison of 4 methods in 2- and 3-dimensional models J Shoulder Elbow Surg (2011) 20, 234-244

glenoidplanefitting.algorithms.plane_fitting.fit_plane_to_points_glenoid(points2, return_meta2=False)[source]

Fit a plane to a set of manually selected points on the glenoid face

Parameters

points1 – np.ndarray, size n by 3 array of the following points, one superior on the glenoid face, two inferior on the glenoid face left and right side
return_meta – If true, also returns the center and normal used to generate the plane

Returns

the fitted plane of the glenoid face

glenoidplanefitting.algorithms.plane_fitting.fit_plane_to_points_scapula(points1, return_meta1=False)[source]

Fit a plane to a set of manually selected points on the scapula

Parameters

points1 – np.ndarray, size n by 3 array of the following points, inferior tip of scapula, medial border of scapula, and center of glenoid fossa.
return_meta – If true, also returns the center and normal used to generate the plane

Returns

the fitted plane through the scapula

glenoidplanefitting.algorithms.plane_fitting.fit_plane_transverse(points1, points3, return_meta3=False)[source]

Fit a transverse plane perpendicular to the scapular plane and passing through the scapular axis.

Parameters

points1 – np.ndarray, size n by 3 array of the following points, inferior tip of scapula medial border of scapula, and center of glenoid fossa.
points3 – np.ndarray, size n by 3 of the following points, center of glenoid fossa, and medial border
return_meta – If true, also returns the center and normal used to generate the plane

Returns

the fitted transverse plane

glenoidplanefitting.algorithms.plane_fitting.planes_version(normal_plane1, normal_plane2)[source]

Determines the glenoid version using the two planes method.

Parameters

normal_plane1 – The normal vector of the scapula plane.
normal_plane2 – The normal vector of the glenoid plane.

Returns

The glenoid version (positive value indicates retroversion)

glenoidplanefitting.algorithms.vault module

This is an implentation of the vault method, see

Matsumura N et al. Computed tomography measurement of glenoid vault version as an alternative measuring method for glenoid version. Journal of Orthopaedic Surgery and Research 2014, 9:17

glenoidplanefitting.algorithms.vault.create_vault_line(anterior_point, posterior_point)[source]

Determines the second point needed to form the Friedman line :param anterior_point: First point on glenoid line, anatomically defined

as a point on the anterior margin of glenoid

Parameters: posterior_point – Second point on glenoid line anatomically defined as a point on the posterior margin of glenoid
Returns: The midpoint of the glenoid line, or the second point of for the vault line

glenoidplanefitting.algorithms.vault.vault_version(anterior_point, midpoint, vaultpoint)[source]

Determines the glenoid version using the glenoid vault as reference :param anterior_point: First point on glenoid line, anatomically defined

as a point on the anterior margin of the glenoid

Parameters

midpoint – Second point on the vault line, anatomically defined as the midpoint of the glenoid fossa
vaultpoint – First point on the vault line, anatomically defined as the tip of the glenoid vault

Returns

The glenoid version (positive value indicates retroversion)

Module contents

glenoidplanefitting.ui package

Submodules

glenoidplanefitting.ui.glenoidplanefitting_command_line module

Command line processing

glenoidplanefitting.ui.glenoidplanefitting_command_line.main(args=None)[source]: Entry point for glenoidplanefitting application

glenoidplanefitting.ui.glenoidplanefitting_demo module

Main entry point function for the various plane fitting functions

glenoidplanefitting.ui.glenoidplanefitting_demo.run_demo(model_file_name, planes='', fried_points='', vault_points='', corr_fried='', output='', visualise=False, config_file=None)[source]

Parameters

planes – File name pointing to file containing points for planes method. First three points are for scapula plane in the order of medial border, inferior tip, glenoid center. For left shoulder latter three points are in order of superior glenoid, left inferior glenoid, right inferior glenoid. For right shoulder latter three points are in order of superior glenoid, right inferior glenoid, left inferior glenoid.
fried_points – File name pointing to a file containing points for the Friedman method. If left shoulder, points in order of, right glenoid tip, left glenoid tip, scapula tip. If right shoulder, points in order of left glenoid tip, right glenoid tip, scapula tip.
vault_points – File name pointing to a file containing points for the vault method. If left shoulder, points in order of, right glenoid tip, left glenoid tip, vault tip. If right shoulder, points in order of, left glenoid tip, right glenoid tip, vault tip.
corr_fried – File name pointing to a file containing points for the corrected Friedman method. Input file: If left shoulder, points in order of, right glenoid tip, left glenoid tip, vault tip. If right shoulder, points in order of, left glenoid tip, right glenoid tip, vault tip.
output – Output filename, can be planes.vtp, friedman.vtp, or vault.vtp Choosing planes.vtp Writes the transverse plane into a file used as the new axial slice for picking the new landmark points for the 3D corrected Friedman method.
config_file – We can pass a configuration file, currently focusing on visualisation parameters

Module contents

glenoidplanefitting

glenoidplanefitting.widgets package

Submodules

glenoidplanefitting.widgets.visualisation module

Widgets for show the results of plane fitting.

glenoidplanefitting.widgets.visualisation.add_vtk_source(renderer, source, linewidth=1.0, opacity=1.0, wireframe=False, colour=None)[source]

simplifies adding a vtk geometry source to a renderer

Parameters

renderer – a vtk renderer to add to
source – a vtk geometry source

glenoidplanefitting.widgets.visualisation.render_window_common(renderer, window_name)[source]

Creates and starts a render window and interactor.

Parameters

renderer – A vtk renderer to add to the render window
window_name – A name for the window

glenoidplanefitting.widgets.visualisation.renderer_common(bone, background_colour=None)[source]: Initialises a vtk renderer and adds the bone model

glenoidplanefitting.widgets.visualisation.vis_fried(bone, cross1, cross2, glenoid1, result, line_width=5)[source]

Visualise the lines resulting from the friedman method.

Parameters

cross2 (cross1,) – The end points of the line crossing the glenoid
result (glenoid1,) – The end points of the line defining the glenoid version

glenoidplanefitting.widgets.visualisation.vis_planes(bone, planes, points1=False, points2=False, resolution=1, plane_size=200.0, vary_plane_colour=False, point_size=5.0)[source]

Visualisation for plane fitting methods

Parameters

bone – The model surface model
planes – a list of planes, as returned by the plane fitting methods in algorithms.plane_fitting

glenoidplanefitting.widgets.visualisation.vis_vault(bone, cross1, cross2, glenoid1, result, line_width=5)[source]

Visualise the lines resulting from the vault method.

Parameters

cross2 (cross1,) – The end points of the line crossing the glenoid
result (glenoid1,) – The end points of the line defining the glenoid version

Module contents

glenoidplanefitting