Issues

• DTI predicts with the measured b=0 but fits without it

  bug
  nipreps/nifreeze#599 · Filter by author oesteban opened on Oct 29, 2026 · M: Models

• Roadmap: registration, models, data, estimator and eddy currents

  documentation
  nipreps/nifreeze#596 · Filter by author oesteban opened on Oct 29, 2026

• Benchmark the memory of the realigned view, to_nifti with motion, and a full run

  enhancement
  nipreps/nifreeze#595 · Filter by author oesteban opened on Oct 29, 2026 · D: Data and memory

• Map NIfTI inputs through an HDF5 cache instead of loading them

  enhancement
  nipreps/nifreeze#594 · Filter by author oesteban opened on Oct 29, 2026 · D: Data and memory

• Tune registration settings on the phantom

  enhancement
  nipreps/nifreeze#593 · Filter by author oesteban opened on Oct 29, 2026 · R2: In-memory registration

• Compare itk-elastix and FireANTs registration backends on the phantom

  enhancement
  nipreps/nifreeze#592 · Filter by author oesteban opened on Oct 29, 2026 · R2: In-memory registration

• Estimate eddy currents as a smooth B-spline along phase encoding

  enhancement
  nipreps/nifreeze#591 · Filter by author oesteban opened on Oct 29, 2026 · R2: In-memory registration

• Register in memory, behind a backend-independent interface

  enhancement
  nipreps/nifreeze#590 · Filter by author oesteban opened on Oct 29, 2026 · R2: In-memory registration

• Keep every cache of a dataset in one directory it owns

  enhancement
  nipreps/nifreeze#589 · Filter by author oesteban opened on Oct 29, 2026 · D: Data and memory

• Three integration tests run ANTs although they mock it

  bug
  nipreps/nifreeze#588 · Filter by author oesteban opened on Oct 29, 2026 · R2: In-memory registration

• to_nifti(insert_b0=True) writes the uncorrected volumes

  bug
  nipreps/nifreeze#587 · Filter by author oesteban opened on Oct 29, 2026 · D: Data and memory

• Test the amplitude of every model's held-out prediction

  enhancement
  nipreps/nifreeze#586 · Filter by author oesteban opened on Oct 29, 2026 · M: Models

• Registration settings from the CLI and from callers are dropped

  bug
  nipreps/nifreeze#585 · Filter by author oesteban opened on Oct 29, 2026 · R2: In-memory registration

• Score registration against the phantom's exact motion

  enhancement
  nipreps/nifreeze#584 · Filter by author oesteban opened on Oct 29, 2026 · R1: Registration harness

• ModelFactory cannot build the PET B-spline model

  nipreps/nifreeze#583 · Filter by author oesteban opened on Oct 28, 2026

• Rerun the sensitivity ladder on every phantom protocol (B7)

  nipreps/nifreeze#575 · Filter by author oesteban opened on Oct 26, 2026

• Run the end-to-end motion test on a phantom (B6)

  nipreps/nifreeze#574 · Filter by author oesteban opened on Oct 26, 2026 · 26.0.0
  1

• Simulate the full phantom grids of every protocol (A5)

  nipreps/nifreeze#568 · Filter by author oesteban opened on Oct 26, 2026

• Move the sensitivity analysis and benchmarks onto the synthetic reference dataset

  enhancement
  nipreps/nifreeze#549 · Filter by author oesteban opened on Oct 22, 2026

• Generate a synthetic reference dataset for the sensitivity benchmarks with TRXScan

  enhancement
  nipreps/nifreeze#548 · Filter by author oesteban opened on Oct 22, 2026
  2

• Select GP hyperparameters by the leave-one-volume-out error

  nipreps/nifreeze#525 · Filter by author oesteban opened on Oct 20, 2026 · M: Models
  1

• Compute the b=0 reference robustly across low-b volumes

  nipreps/nifreeze#520 · Filter by author oesteban opened on Oct 20, 2026
  1

• Repair the documentation notebooks (the weekly run has failed since 2025-09)

  nipreps/nifreeze#519 · Filter by author oesteban opened on Oct 20, 2026
  1

• Fit models from a masked, time-first cache in voxel blocks

  nipreps/nifreeze#513 · Filter by author oesteban opened on Oct 20, 2026 · D: Data and memory
  1

• GQI leave-one-volume-out predictions are overscaled

  nipreps/nifreeze#495 · Filter by author oesteban opened on Oct 19, 2026 · M: Models
  1