Skip main navigation

Femur project: first step

First steps in your Scalismo shape modelling project: reconstructing partial femur shapes.
© University of Basel

You are now ready to start working on your own shape modelling project: reconstructing partial femur shapes.

Please note that this project is optional. You can continue your learning journey through this course even if you decide not to do the project.
Giving it a try will however help you assess your understanding of the concepts you have learned and use them in a real-world application. In case you decide to do the project, please make sure to follow the full track version of the Scalismo Lab tutorials.

Project overview:

This project will last until the end of this course and will proceed in several steps:

  1. You will build a femur model using only one provided femur mesh as reference and the Gaussian Process modelling skills you have acquired so far.

  2. The following week, you will access an online medical imaging repository (the Sicas Medical Image Repository (SMIR)) and download a set of femur meshes and corresponding landmarks that you will need for building a statistical shape model.

  3. At the end of Week 6, you will use the femur model you have built in the first step and fit it to the dataset prepared in the second step. The goal is to establish correspondence and build a statistical femur model from data. This model will then be used to reconstruct missing parts in a provided dataset of incomplete femurs.

To spice up things and make the project a bit more fun, we also host a competition for the best shape reconstructions on SMIR. Once your femur reconstructions are complete, you will have the opportunity to upload them to the repository and compare them to the ground truth, that is the original complete shape.

The uploaded reconstructions will then be ranked based on their distance to the ground truth, allowing you to know how well you performed compared to your fellow learners.

More details on this competition will be provided in the third phase of the project, in two weeks.

Create your account on SMIR:

To create your account on SMIR, follow the procedure below :

  • Go to the SMIR registration page.

  • Fill in your details, and select SSM.FUTURELEARN.COM as a research unit.

  • This will send a request to an administrator to authorize your account creation. Please bare in mind that this might take up to 24h. You will be informed by Email once your account creation is authorized.

We will use this account in the second step, next week, to download the femur data.

We are now at the first step:

As mentioned above, the goal in this step is to build a femur model.

To do so, proceed as follows:

  • Locate the provided reference femur mesh, shipped with the Scalismo Lab tutorial package under datasets/femur.stl

  • Use the shape modelling and Scalismo concepts you have learned so far to build a Gaussian Process yielding smooth deformations and combine it with the provided reference mesh to build a StatisticalMeshModel of the femur.
    Important: visualisation is very crucial at this stage. Make sure to visualise the shapes resulting from your model and verify that it exhibits the right amount of flexibility.

  • Once you are satisfied with the generated StatisticalMeshModel, store it to file using the StatismoIO.writeStatismoMeshModel method or by displaying it and saving it via the GUI (right-click on the model name in the scene)

That’s it! If you have done all this, you have completed the first step of the project.

© University of Basel
This article is from the free online

Statistical Shape Modelling: Computing the Human Anatomy

Created by
FutureLearn - Learning For Life

Reach your personal and professional goals

Unlock access to hundreds of expert online courses and degrees from top universities and educators to gain accredited qualifications and professional CV-building certificates.

Join over 18 million learners to launch, switch or build upon your career, all at your own pace, across a wide range of topic areas.

Start Learning now