Statistical Shape Modelling: Computing the Human Anatomy

Learn modern methods that will help shaping the future of medical interventions

Statistical shape models are one of the most important technologies in computer vision and medical image analysis. With this technology, the computer learns the characteristic shape variations of an object or organ. The model resulting from this analysis may then be used in implant design, image analysis, surgery planning and many other fields.

In this free online course, you will get insights from mathematics, statistics and machine learning, in order to address practical problems, as well as a theoretical and practical introduction to the open source software Scalismo.

This course is intended for students and professionals with a Bachelor in computer science, medical imaging professionals and biological anthropologists, who are interested in top-notch research, scientific insights and a useful application. Please note that tutoring usually takes place between End of February and Mid May. The course addresses people with some previous knowledge in computer science. If you find it challenging it would be a good idea to wait for the next mentoring window.

Although you can watch the videos, read the articles, and complete the tests and quizzes on mobile devices such as smartphones or tablets, you will have to install the free software Scalismo on your own workstation in order to use it – there is no online version available.

In order to be able to do this, your computer should meet the following minimum system requirements: Windows (32bit/64bit), Mac OS X or Linux (64bit), 4GB of RAM, 500MB of free HD space. There are no special requirements for the graphic adapter.

To take part in this course, you need to download and install Scalismo Lab and create your account on SMIR.

• Basic concepts of shape modelling and the Scalismo software
• Course introduction
• Shape modelling basics
• Introducing the software environment
• Building shape models: fundamental concepts in theory and practice
• Modelling shape variations using Gaussian Processes
• Experimenting on a face dataset using Scalismo
• Probabilistic aspects of shape models
• Shape models and the normal distribution
• Probabilistic aspects of shape models in Scalismo
• Parametric representations of shape models
• Modelling with Gaussian Processes
• Modelling with kernels
• Modelling with kernels in Scalismo
• Flexibility of statistical shape models
• Half-time!
• Reconstructing missing parts
• Shape modelling in medical practice
• Incorporating known deformations into shape models
• Gaussian Process regression in Scalismo
• Femur project
• Fitting models to data
• The fitting problem
• Iterative Closest Points in Scalismo
• Femur project
• Model-based image analysis
• Fitting shape models to images
• Fitting images in Scalismo
• Femur project: progress
• Conclusion and outlook
• Applications and future directions
• Femur project
• Summary and outlook