Skip to content

CHANGELOG.md

Latest commit

 

History

History
289 lines (207 loc) · 15.5 KB

CHANGELOG.md

File metadata and controls

289 lines (207 loc) · 15.5 KB

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

Unreleased

0.4.1 - 2026-06-16

Added

  • multiple corresponding units for given quantities in the GUI

Fixed

  • missing keys in init_beam not being skipped
  • hotfix for deprecated numpy.matrix in to_namespace()

0.4.0 - 2026-05-17

Added

  • pyRadPlan.ai_agents module: LLM-powered treatment planning helpers built on pydantic-ai
  • generate_beam_angles(pln, treatment_site) — queries an LLM to suggest gantry and couch angles for a given treatment site and radiation mode, and writes them into pln.prop_stf
  • generate_voi_objectives(pln, cst, treatment_site) — queries an LLM to propose optimization objectives for each VOI in a StructureSet and attaches validated objective instances directly to the VOIs
  • AiSettings — pydantic-settings class for global configuration; reads the default model from the PYRADPLAN_AI_MODEL environment variable (default: claude-sonnet-4-5); any provider supported by pydantic-ai can be selected via the model string
  • New example examples/utils_ai_agents.py demonstrating an end-to-end proton prostate plan with AI-generated beam angles and objectives
  • AGENTS.md and CLAUDE.md for AI-assisted development
  • GPU-accelerated dose calculation via Array API using CuPy and PyTorch backends (alongside NumPy/array_api_strict), including memory management, streaming, and per-beam cleanup
  • to_namespace() helper to convert arrays (and scipy sparse matrices) between Array API namespaces, with explicit device= and keep_sparse_compat options
  • choose_device() to select a sensible default device for a given namespace, with multi-GPU index support (gpu:N / cuda:N)
  • DLPack-based device handling (get_device_info, is_on_gpu, DLPACK_CPU/DLPACK_CUDA constants) for seamless backend interop
  • GPU lifecycle helpers: free_gpu_memory(), create_stream(), get_current_stream(), synchronize(), record_event(), elapsed_time()
  • from_numpy() / to_numpy() helpers with device targeting
  • Backend availability checks and a preferred-backend wishlist (cupy_available, pytorch_gpu_available, jax_available, jax_gpu_available, numba_cuda_available, PREFERRED_GPU_ARRAY_BACKEND)
  • Native CUDA kernel implementation for geometric distance calculation: _calc_geo_dists_cupy_kernel (CuPy ElementwiseKernel), _calc_geo_dists_cupy_raw_kernel (CuPy RawKernel), and _calc_geo_dists_torch_kernel, replacing the previous Numba CUDA path
  • Array API conform N-D interpolation interpnd() on rectilinear grids (generic 2D/3D fallback, with dedicated RegularGridInterpolator paths for NumPy/SciPy, JAX, and CuPy)
  • Improved interp1d: fast paths using xp.interp for NumPy/JAX/CuPy, JAX jit and PyTorch torch.compile backends (with torch.jit.script fallback if Triton is not installed), support for lists/tuples/dicts of arrays with optional stacking
  • Array API compatibility for beam initialization and ray geometry computation (get_gantry_rotation_matrix, get_couch_rotation_matrix)
  • More efficient sparse matrix conversion using direct CSR/CSC construction, avoiding unnecessary deep copies
  • Kernel-data caching and ParticlePencilBeamKernel.to_namespace() to move kernel arrays onto the active backend/device
  • Device propagation through fluence optimization (NonLinearFluencePlanningProblem, OptimizerIpopt, scipy solver, SolverBase.device)
  • New example examples/utils_backends.py demonstrating how to query backends and run dose calculation on different array backends
  • Benchmarks benchmark/benchmark_interp1d.py and benchmark/benchmark_interpnd.py for interpolation across backends
  • quantity resolver that checks for presence of quantities and instantiates the required ones
  • biological RBE calculation from alpha and beta kernels
  • biological based optimization
  • alpha and beta parameters to dij with function get_reference_lq_params to get them for a given ct and cst
  • quantity resolver that checks for presence of quantities and instantiates the required ones
  • a TOPAS monte carlo interface. Implemented are protons and ions with basic physical and let based scoring. Material conversions are water and a pre defined schneider converter. The interface is structure into input files so that different beam models or scorers can be added easily. The interface also as template files that are used to create the simulation files. Which also give a nice overwiev on the structure of the simulation files.
  • jinja2 as project dependency
  • Documentation: extended user guide
  • Documentation: extended installation instructions
  • First implementation of an interactive result viewer widget (gui extra) for visualizing dose/quantity distributions slice-by-slice with scroll/zoom, VOI contour overlay, isoline rendering, and colormap selection
  • DVH viewer and DVH comparison tools in the GUI analysis widget
  • Quality indicator (QI) panel in the GUI analysis widget
  • visible and visible_color fields on VOI for storing per-structure display properties
  • DEFAULT_VOI_COLORS palette (per VOI type: TARGET, OAR, EXTERNAL, HELPER) exported from pyRadPlan.cst
  • StructureSet.set_colors() method that auto-assigns colors from the predefined palette, skipping colors already in use, preserving any explicitly set visible_color
  • visible_color field validator on VOI accepting named color strings, float 0–1 arrays, and int 0–255 tuples

Changed

  • Pencil beam dose calculation now applies the lateral cutoff mask before computation rather than after, and keeps Dij assembly on the CPU (size limited) while the rest of the calculation runs on GPU
  • Per-beam dose/LET/effect computation in Dij.compute_beam_dose() now slices intensities by beam (faster matmul) instead of multiplying by a beam mask, and is fully Array API namespace aware
  • Siddon raytracer now picks up the engine's device and allocates plane/coordinate arrays directly on that device
  • SVD photon pencil beam engine refactored to use Array API arrays for ray-position aggregation and kernel weighting (still calls SciPy interpolators on host arrays)
  • resampling in BLD now uses interpolation to return mask on the grid provided by the dose engine
  • in SVDPB field_grid is now built before the resampling of the beamlet mask to guarantee matching grids
  • _draw_contours() in plot_slice now uses voi.visible_color instead of a colormap cycle, so contour colors are consistent with the GUI viewer
  • GUI optional dependency changed from PyQt5 to pyside6>=6.0.1 and pyqtgraph>=0.12.0
  • Proton pencil-beam example updated to use the new dose viewer widget

Fixed

  • resample_image now falls back to linear interpolation when BSpline is requested but the image has fewer than 4 voxels in any dimension, preventing intermittent NaN values from the BSpline prefilter on small grids
  • CuPy issue in LPS coordinate handling (gantry/couch rotation matrices now built via xp.stack with the correct device/dtype)
  • SVD pencil beam engine updated to match changes in the base pencil beam engine
  • Device handling and type checks in optimization solvers (IPOPT and SciPy) so the optimization runs on the same device as the quantities
  • Preliminary workaround for CUDA / cuBLAS DLL conflicts when PyTorch and CuPy are imported in the same environment
  • free_gpu_memory() now skips NumPy/array_api_strict namespaces silently
  • Beam.validate_nparray_dtype handles non-list array inputs via to_numpy() so non-NumPy arrays validate correctly on import
  • to_namespace() raises TypeError for scalar / list / tuple inputs instead of failing on the sparse-array check
  • StructureSet now calls set_colors() during validation so every VOI always has a color assigned
  • property .size raising an error in dij.py when torch is used. Switching to array_api_compat.size()

0.3.5 - 2026-05-12

Changed

  • transmission mask in beam limiting devices now uses edge smoothing to account for finite grid size

Fixed

  • Bugfix in FRED MC engine, where temporary files were not deleted

0.3.4 - 2026-05-05

Added

  • dose engines now have private flags _dij_guarantee_canonical and _dij_guarantee_nonzero to guarantee sparse dij structure before finalization
  • Issue and Pull/Merge Request templates
  • default pre-commit hooks for line endings, case conflicts, merge conflicts, etc. Also added codespell, CITATION.cff verifyer and toml check
  • plot_slice() and plot_multiple_slices() now accept a generic image_volume (CT, dict, or sitk.Image) instead of only a CT via the ct argument
  • plot_slice() gained image_window and window_mode ("minmax" or "centerwidth") options for grayscale windowing, replacing the previous ct_window
  • overlay_unit in plot_slice() now also accepts a pint.Quantity
  • Jennifer Josephine Hardt added as author (CITATION.cff and pyproject.toml)
  • Added an extrapolate option to resample_image to choose the extrapolator. Default is to use a nearest neighbor extrapolator, but individual values can beprovided as well.

Changed

  • New version of photons_Generic.mat basedata file can now be provided, allowing a "version" field alongside "meta" and "data" files within the machine struct. Version 2 requires correct kernel normalization (without implying a spacing in the convolution integral). photons_Generic.mat has been updated to version 2 with correct kernel normalization.
  • Photon dose calculation now does not rely on hardcoded convolution resolution integral normalization of machine kernels. Assumes that old kernels use hardcoded factor of 4 for 0.5 mm resolution (1/0.5^2).
  • Sparse structure now uses one shared index structure across dijs if possible
  • plot_slice() internally refactored into modular helpers (input validation, axis resolution, image/contour/overlay drawing, scale bar, finalization) for readability and reuse
  • Slice titles now follow the format Slice ID = N at X mm (or Slice ID = N when no image volume is provided)
  • plot_slice()/plot_multiple_slices() parameter ct renamed to image_volume and ct_window renamed to image_window

Deprecated

  • ct and ct_window arguments to plot_slice()/plot_multiple_slices() still work as aliases for image_volume and image_window but emit a DeprecationWarning

Fixed

  • CI Release Workflow now tests correctly on release commits without an Unreleased section in the Changelog.
  • CI Release Workflow now fetches tags correctly and supplies the correct release body from the tag message

0.3.3 - 2026-04-14

Added

  • Automatic Release workflow on GitHub reading CHANGELOG.md and tag message
  • Benchmark Folder with initial Raytracer benchmark that can be run with pytest-benchmark

Changed

  • Changelog now follows thee Keep A Changelog conventions

Fixed

  • Fixed raytracer issue where certain geometrical configurations could lead to individual rays starting with invalid indices, inserting misplaced NaN's into the radiological depth cube.

0.3.2 - 2026-04-07

Added

  • Mimicking objective for dose-mimicking optimization (SquaredMimicking)
  • Prototype for field-based dose calculation using Beam Limiting Device & FieldShape hierarchy
  • CITATION.cff with authors and ORCIDs

Fixed

  • Raytracer candidate matrix now uses lateral cutoff by default (overridable)
  • Fixed binary mask interpolation in VOI
  • Fixed import of empty voxel index lists from matRad
  • Fixed crash when optimization problem has no objectives
  • Fixed export of None values and type check in dij/grids
  • Fixed missing x-divergence of ray in FRED MC engine
  • Fixed single-field STF generator to properly inherit from IMRT generator

0.3.1 - 2026-01-28

Added

  • Performant and Array API compatible candidate ray matrix setup alternative for cube raytracing
  • Convenience plotting function to display multiple slices (plot_distributions)

Changed

  • numba is no longer a mandatory dependency

Fixed

  • Fix readthedocs YAML to correct Python version
  • Small code quality fixes

0.3.0 - 2026-01-12

Added

  • Partial Array API compatibility with GPU support for CuPy and Torch (drops Python 3.9)
  • FRED interface (Monte Carlo tool)
  • VHEE planning with a generic (unfocused) beam and a focused beam
  • dij.compute_result_ct_grid() now returns quantities per beam
  • create_body_segmentation() method for the CST object
  • Option to cancel solver at any iteration via keyboard input
  • Comprehensive Sphinx documentation
  • Various examples conforming to jupytext norm

Changed

  • Increased memory efficiency in dose calculation
  • Tuned initial Scipy solver parameters
  • Refactored cst, ct, machine, and stf test data
  • Added Python version matrix to CI tests

Fixed

  • Fixed overlap priorities when similar levels exist
  • Fixed np.floating deprecation
  • Fixed pydantic >= 2.11 compatibility
  • Fixed issues with single bixel calculations in raytracer
  • Elevated minimum required version of numpydantic

0.2.8 - 2025-06-28

Added

  • DVHCollection and DVH for plan analysis
  • Maps to associate bixel/beamlet indices with beams/rays in stf

Changed

  • Performance improvements for raytracer and dij filling

Fixed

  • Ray tracer recovery in case of numerical issues
  • CT validates given x/y/z vectors correctly
  • Fixed validation of VOIs with single voxels imported from matRad

0.2.7 - 2025-06-27

0.2.6 - 2025-06-26

0.2.5 - 2025-06-20

0.2.4 - 2025-05-22

0.2.3 - 2025-05-09

Added

  • Slice visualization function
  • LET calculation for protons

Changed

  • Scenarios are now pydantic models
  • Docstring and code quality improvements

Fixed

  • Performance fix for raytracer
  • Fixed issues with ray validation
  • Various validation fixes

0.2.2 - 2025-02-27

Added

  • matRad-compatible data structures with stable validation and serialization using pydantic
  • Native reimplementation of matRad's pencil beam dose calculation for photons, protons & ions
  • Generic machine data
  • Native optimization framework using scipy or IPOPT (via ipyopt)

0.2.1 - 2025-02-20

0.2.0 - 2025-02-13

0.1.0 - 2025-02-05

0.0.2 - 2025-01-10

0.0.1 - 2025-01-10