Announcing the First Official Release of OpenTPS!

General
1

The OpenTPS team is thrilled to announce the very first official release of OpenTPS! Our goal is to enhance version tracking for the software and address bugs effectively. Each release will include detailed release notes outlining resolved issues and improvements. You can view the release notes for the current version (1.1.2) below.

We welcome your feedback—please don’t hesitate to report any bugs or suggest new features.

1 Like
2

OpenTPS v1.1.2 release notes :

New Features

[Processing/Optimization]

  • Added new objective types in optimization: DUNIFORM, DVHMIN, DVHMAX, EUDMIN, EUDMAX, EUDUNIFORM, DFALLOFF

  • Integrated new SciPy solvers for optimization: Scipy-SLSQP, Scipy-COBYLA, Scipy-trust-constr.

Bug Fixes

[IO]

  • Resolved data export issue to DICOM files.

[Processing]

  • Fixed convergence issue in getConvergenceData function.

  • Modified C code for energy layer placement.

  • Corrected scenario computation in OpenTPS, aligning with MCsquare. Renamed scenario selection strategies: ALL, REDUCED, RANDOM.

  • Ensured full-cupy morphons implementation compatibility with different spacings and origins between fixed and moving images.

  • Updated old image attributes used in WET raytracing.

[Data]

  • Corrected ScoringGridSize setter in plan design.

  • Fixed issue of duplicate nominal scenario compuration in MCsquare and openTPS-core

[GUI]

  • Fixed GUI issue where the patient was not visible on a slide.

Improvements

[Processing]

  • Enhanced support for handling multiple target volumes for spot placement.

  • Stopped saving dose in patient by default when optimizing to avoid it to appear twice in GUI.

  • Save the index of the worst scenario to avoid a recomputation in gradient.

[Examples]

  • Resolved “Local path in dose calculation example” (hardcoded path):

  • Added a small water phantom dataset in testData/ folder.

  • Rewritten example to demonstrate reading DICOM files (CT, plan, dose & contours) and dose computation with MCsquare engine.

  • Add examples to support multiple environments when running a script

[Data]

  • Renamed Robustness class to RobustnessEval and created a new Robustness class specific for optimization. Reason: Robustness settings for optimization may not be reused for evaluation (numScenario, etc.).

  • Adding PEHD to mcsquare materials

  • Read rigid registration dicom file

Other Changes

  • Fixed issue with latest Matplotlib version (3.9.0).

  • Added a message box when opening the GUI indicating that OpenTPS is not a medical device.

1 Like