| | | | Define ShapeWorks Future Roadmap!
Our vision is to make ShapeWorks a standard tool for shape analyses in medicine. Your input matters to realize this vision! | | | |
|
| | | | | It only takes one minute to help defining ShapeWorks future. | | | |
|
|
|
|
|
| | | | ShapeWorks Users Get-Together | | | |
|
|
|
| | | | We would like to cordially invite you to our next remote ShapeWorks monthly users get-together. This get-together will be an open discussion to hear your estimable feedback and recommendations about Shapeworks 6.1 and features that you have been using and/or would like to have in ShapeWorks.
ShapeWorkers are here to answer your questions and turn your feedback into actionable items. This discussion is also an excellent opportunity to bring your shape modeling story on the table to discuss how we could help you customize the released use cases to your data and study. | | | |
|
|
|
| | | | In case you missed our workshop @ SB3C 2021 … | | | |
|
|
|
| | Introduction to Shape Modeling & ShapeWorksThe first part of the workshop focuses on basics and theory of shape modeling. We also demonstrate ShapeWorks Studio, the all-in-one user-friendly tool for interactive anatomy modeling. | |
|
| |
|
|
|
| | ShapeWorks Success Stories in Orthopedics
In this part, we highlight the robust utility of ShapeWorks in orthopedics-related clinical applications that hinge on the quantitative study of anatomical forms. | |
|
| |
|
|
|
| | Next Generation of ShapeWorks
In the last part of the workshop, we focus on the next generation of ShapeWorks including advanced and state-of-the-art shape modeling features and where we want to be in the next years. | |
|
| |
|
|
|
| | What is new in ShapeWorks 6.1? | | | |
|
|
|
| | Multiple Domains Support & More ...Multiple domains support in Studio: Multiple domains are added to Studio where shape models are optimized in the given domains' shared/joint shape spaces to capture inter-domains correlations and interactions. The way multiple domains are implemented allows for an arbitrary number of domains as long as all shape samples in a given cohort have the same domains. See Multiple Domains SSM for more details.
Grooming support for meshes in Studio: Multiple grooming features for mesh domains are added to Studio, including mesh smoothing, hole filling, and rigid alignment with auto reference selection. See New in ShapeWorksStudio for more details.
Python APIs for shape statistics: Python APIs for principal component analysis (PCA) are added for shape statistics. See ellipsoid_pca use case for a demonstrating example. | |
|
| |
|
|
|
| | | Improved Models, Contours Support & More ...Improved shape models for convoluted structures: Geodesic distance-based repulsion is now supported for mesh domains. This improves results on structures with thin and/or convoluted features at the cost of increased memory and runtime. Refer to the thin_cavity_bean use case for an example. Optimizing shape models on contours: N-dimensional contour domains are now supported in ShapeWorks. Refer to the supershapes_1mode_contour use case for an example. Robust and scalable primitive-based constrained surface sampling: Constraints are now implemented by turning the problem into an unconstrained optimization using the augmented lagrangian inequality formulation. The new implementation supports both cutting planes and spheres. It also supports multiple constraints per domain/shape with different types. Consolidated library for shape statistics: The shapeworks computational library is updated to include refactored code for shape statistics, including functions to read particle files and compute eigenvectors, eigenvalues, and PCA loadings. | |
|
|
|
|
|
| | Refactors and Improved DeepSSM & More ...Refactored DeepSSM: Defined a config file for DeepSSM parameters that are used in training and testing. This will be helpful for parameter tuning, model comparison, and adding additional functionality and parameters to DeepSSM. DeepSSM with fine-tuning: A fine-tuning option has been added to DeepSSM. This allows the model to learn the mapping between the PCA space to the correspondence point space, improving accuracy. Improved DeepSSM evaluation: DeepSSM evaluation has been adapted to use ShapeWorks mesh warp function and Python binding mesh distance function. DeepSSM on both GPU and CPU: The DeepSSM use case has been adapted to run on both GPU and CPU. It is considerably faster on GPU but no longer exits when running on CPU. | |
|
| |
|
|
|
| | | New Use Cases, Improved Notebooks & More ...Multi-domain shape cohort generation: Ellipsoid joint generation is added to ShapeWorks' cohort generator python module to showcase and troubleshoot multiple domain shape modeling workflows. New use cases: New use cases are added to demonstrate PCA tools, geodesic distance-based particle-to-particle interactions, modeling contours, and multiple domains SSM. See Examples/Python
Improved documentation: More clear documentation both from Python, the command line, and online are added.
Improved notebooks: Notebooks are significantly simplified by moving helper functions into the ShapeWorks Python module, reducing redundant information, and using only the PyVista library for visualization. | |
|
|
|
|
|
| | Installation Improvements- Improved installation: Users no longer are required to modify PATHs or pass parameters to find executables or import the ShapeWorks Python module. Installation works even if non-standard installation directories are used and it works on all platforms.
- New shapeworks environments can now be created using "install_shapeworks [name]": This supports multiple installations on the same platform, where users can change installation just by activating a different conda environment.
- Clean installation for ShapeWorks Python module: ShapeWorks Python module is now seamlessly installed as part of ShapeWorks suite installation on all platforms. The Python module no longer requires any special path modifications to import.
| |
|
|
|
| | Discussion forum: Join ShapeWorks' discussion forum to post questions and comments. ShapeWorks dev team is more than happy to help troubleshoot issues that you might face when using ShapeWorks on your data.
Email list: To join our mailing list, kindly send a message to sympa@sci.utah.edu with the subject subscribe shapeworks-users and an empty body.
Reporting bugs and feature requests: Please feel free to submit feature requests and report bugs on ShapeWorks GitHub issue tracker and we will add more discussions on the issue as we do progress on this. This will help us prioritize this feature as an action item.
| |
|
|
|
|
