v4.0.0

v4.0.0 — clean-slate redesign

A ground-up redesign of the package: same algorithms (better tested), new interface.

• One CLI: eleven scripts become subcommands of intensity-normalize, with unified single-image / batch semantics.
• New Python API: individual methods are plain functions (inorm.whitestripe(img, mask=mask)); population methods are fitted transforms (tx = inorm.nyul.fit(imgs); tx(new_img); tx.save("nyul.npz")).
• Type-preserving: numpy in → numpy out, nibabel in → nibabel out (affine/header preserved, float64 kept float64).
• Array-level cores for every method: pure numpy, no nibabel, for use inside your own pipelines.
• Deterministic by default: all stochastic steps take seed= (default 0); scikit-fuzzy replaced by an in-house fuzzy c-means.
• Saved transform state moves from .npy to stamped .npz (old files are not loadable — refit them).
• Requires Python ≥ 3.12; pymedio support removed.

Migration guide: https://www.jcreinhold.com/intensity-normalization/migration/
Docs: https://www.jcreinhold.com/intensity-normalization/

3.0.1

Add fix by @ndonyapour to KDE

3.0.0

Overhaul for simplicity and maintainability

v2.2.4 - Lauterbur

• Allow Python >=3.9

v2.2.3 - Lauterbur

• Revert error on different image shapes from RavelNormalize; it is required!

v2.2.2 - Lauterbur

• Remove plural from Modality and TissueType enumerations.
• Update tutorials to use the Modality and TissueType enumerations.
• Remove error on different image shapes from RavelNormalize when registration enabled.

v2.2.1 - Lauterbur

• Update documentation to support modifications to Python API
• Update dependencies
• Remove incorrect warning from WhiteStripe normalization

v2.2.0 - Lauterbur

• Change backend to pymedio to support more medical image formats

v2.1.4 - Austin

• Fix testing bugs in 2.1.3 and cleanup some interfaces

v2.1.2 - Austin

• Improve security