Fakultät Informatik und Mathematik
Regensburg Center for Artificial Intelligence
Regensburg Center of Biomedical Engineering
Regensburg Center of Health Sciences and Technology

Prof. Dr. rer nat. Christoph Palm

Regensburg Breast Shape Model (RBSM)

 

The Regensburg Breast Shape Model (RBSM) is a 3D statistical shape model of the female breast, learned from 110 breast scans acquired in a standing position. It is the first which is publicly available and free for research purposes. Read more about the RBSM in our paper:

Maximilian Weiherer, Andreas Eigenberger, Bernhard Egger, Vanessa Brébant, Lukas Prantl, Christoph Palm.
Learning the shape of female breasts: an open-access 3D statistical shape model of the female breast built from 110 breast scans.
Vis Comput (2022). https://doi.org/10.1007/s00371-022-02431-3

This work was done in cooperation between the Regensburg Medical Image Computing (ReMIC) lab at OTH Regensburg and the University Center of Plastic, Aesthetic, Hand and Reconstructive Surgery at the University Hospital Regensburg.

 

Download the RBSM


The RBSM is available in the HDF5 file format as provided by the Scalismo framework. Besides the full model build from all 110 breast scans, we also provide the following bespoke models trained only on a subset of our database:

  • an operated-only model containing the 72 operated breasts,
  • a non-operated-only model containing the 38 non-operated breasts,
  • a model built from participants under 30 years of age (30 breasts),
  • a model built from participants between 30 and 50 years of age (50 breasts),
  • a model built from participants older than 50 years of age (30 breasts),
  • a model trained on participants with an absolute volume difference between left and right breast less than 50cc (52 breasts),
  • a model trained on participants with an absolute volume difference between left and right breast between 50 and 150cc (40 breasts),
  • a model trained on participants with an absolute volume difference between left and right breast greater than 150cc (18 breasts),
  • a model built from participants with a BMI less than 25 (83 breasts),
  • a model built from participants with a BMI greater than 25 (27 breasts).

NEW! We built a second collection of models trained on an augmented dataset which also includes the mirrored versions of the 110 breast scans in our original database, leading to 220 breast scans in total. We provide the same bespoke models as above.

To download the RBSM, please carefully read the following license agreement first:

  • NON-COMMERCIAL USE: The license granted is for internal, non-commercial research, evaluation or testing purposes only. Any use of the RBSM or its contents to manufacture or sell products or technologies (or portions thereof) either directly or indirectly for any direct or indirect for-profit purposes is strictly prohibited.
  • NO WARRANTY: The RBSM is provided as is and any express or implied warranties are disclaimed.
  • NO DISTRIBUTION: You are not allowed to sublicense or distribute the RBSM in whole or in part to any third party.
  • CITATION: You agree to reference the publication

Maximilian Weiherer, Andreas Eigenberger, Bernhard Egger, Vanessa Brébant, Lukas Prantl, Christoph Palm.
Learning the shape of female breasts: an open-access 3D statistical shape model of the female breast built from 110 breast scans.
Vis Comput (2022). https://doi.org/10.1007/s00371-022-02431-3

 

RBSM-110

Visualize the RBSM


The RBSM can be easily visualized using our custom, python-based RBSM Viewer. We provide an installer as well as a zipped executable here:


We currently support only Windows platforms (Windows 7, 8, 10, and, of course, the good old Windows XP). However, we are actively working on a version for MacOS!


Besides, our model can be inspected using the Scalismo Viewer, developed by the Graphics and Vision Research Group of the University of Basel. We provide a pre-compiled, executable jar file here:

Note that Java need to be installed on your computer (Java version 8.0 or higher is recommended).

Loading the RBSM into the Scalismo Viewer: Open the Scalismo Viewer -> Right click on Default Group -> Select Load Statistical Shape Model -> Browse for the downloaded RBSM and click Open.

A pictured step-by-step guide can be found here.


In case you have any questions, please feel free to contact Maximilian Weiherer (maximilian.weiherer@fau.de) or Prof. Dr. Christoph Palm (christoph.palm@oth-regensburg.de).

 

 

The Regensburg Breast Shape Model (RBSM) is a 3D statistical shape model of the female breast, learned from 110 breast scans acquired in a standing position. It is the first which is publicly available and free for research purposes. Read more about the RBSM in our paper:

Maximilian Weiherer, Andreas Eigenberger, Bernhard Egger, Vanessa Brébant, Lukas Prantl, Christoph Palm.
Learning the shape of female breasts: an open-access 3D statistical shape model of the female breast built from 110 breast scans.
Vis Comput (2022). https://doi.org/10.1007/s00371-022-02431-3

This work was done in cooperation between the Regensburg Medical Image Computing (ReMIC) lab at OTH Regensburg and the University Center of Plastic, Aesthetic, Hand and Reconstructive Surgery at the University Hospital Regensburg.

 

Download the RBSM



The RBSM is available in the HDF5 file format as provided by the Scalismo framework. Besides the full model build from all 110 breast scans, we also provide the following bespoke models trained only on a subset of our database:

  • an operated-only model containing the 72 operated breasts,
  • a non-operated-only model containing the 38 non-operated breasts,
  • a model built from participants under 30 years of age (30 breasts),
  • a model built from participants between 30 and 50 years of age (50 breasts),
  • a model built from participants older than 50 years of age (30 breasts),
  • a model trained on participants with an absolute volume difference between left and right breast less than 50cc (52 breasts),
  • a model trained on participants with an absolute volume difference between left and right breast between 50 and 150cc (40 breasts),
  • a model trained on participants with an absolute volume difference between left and right breast greater than 150cc (18 breasts),
  • a model built from participants with a BMI less than 25 (83 breasts),
  • a model built from participants with a BMI greater than 25 (27 breasts).

NEW! We built a second collection of models trained on an augmented dataset which also includes the mirrored versions of the 110 breast scans in our original database, leading to 220 breast scans in total. We provide the same bespoke models as above.

To download the RBSM, please carefully read the following license agreement first:

  • NON-COMMERCIAL USE: The license granted is for internal, non-commercial research, evaluation or testing purposes only. Any use of the RBSM or its contents to manufacture or sell products or technologies (or portions thereof) either directly or indirectly for any direct or indirect for-profit purposes is strictly prohibited.
  • NO WARRANTY: The RBSM is provided as is and any express or implied warranties are disclaimed.
  • NO DISTRIBUTION: You are not allowed to sublicense or distribute the RBSM in whole or in part to any third party.
  • CITATION: You agree to reference the publicationMaximilian Weiherer, Andreas Eigenberger, Bernhard Egger, Vanessa Brébant, Lukas Prantl, Christoph Palm.
    Learning the shape of female breasts: an open-access 3D statistical shape model of the female breast built from 110 breast scans.
    Vis Comput (2022). https://doi.org/10.1007/s00371-022-02431-3

 

RBSM-110

Visualize the RBSM


The RBSM can be easily visualized using our custom, python-based RBSM Viewer. We provide an installer as well as a zipped executable here:


We currently support only Windows platforms (Windows 7, 8, 10, and, of course, the good old Windows XP). However, we are actively working on a version for MacOS!


Besides, our model can be inspected using the Scalismo Viewer, developed by the Graphics and Vision Research Group of the University of Basel. We provide a pre-compiled, executable jar file here:

Note that Java need to be installed on your computer (Java version 8.0 or higher is recommended).

Loading the RBSM into the Scalismo Viewer: Open the Scalismo Viewer -> Right click on Default Group -> Select Load Statistical Shape Model -> Browse for the downloaded RBSM and click Open.

A pictured step-by-step guide can be found here.


In case you have any questions, please feel free to contact Maximilian Weiherer (maximilian.weiherer@fau.de) or Prof. Dr. Christoph Palm (christoph.palm@oth-regensburg.de).