Changelog¶

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

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

[0.1.0] - Unreleased¶

Initial release of qmri, providing pure MRI signal models and fitting algorithms for quantitative MRI analysis.

Diffusion imaging
- ADC (Apparent Diffusion Coefficient) fitting with multiple methods:
  - LLS (Linear Least Squares)
  - WLLS (Weighted Linear Least Squares)
  - IWLLS (Iterative Weighted Linear Least Squares)
- DWI signal generation from ADC values
- Support for single voxel and volumetric data
Relaxometry
- T1 mapping with multiple acquisition methods:
  - IR (Inversion Recovery with finite TR correction)
  - IR Classical (Inversion Recovery assuming TR >> T1)
  - VTR (Variable TR for spoiled gradient echo)
- T2 mapping from multi-echo spin echo data:
  - Full model (with offset term)
  - Reduced model (pure mono-exponential)
- Signal generation functions for both T1 and T2
Perfusion
- ASL (Arterial Spin Labelling) quantification
- GKM (General Kinetic Model) for DCE-MRI
Thermometry
- PRF (Proton Resonance Frequency) shift method for MR thermometry
Fitting utilities
- Least squares algorithms
- Bootstrap uncertainty estimation
Error metrics
- R-squared (coefficient of determination)
- RMSE (Root Mean Square Error)
- Uncertainty propagation
Constants
- Physical constants for MRI calculations