Clinica 0.4.0 - PET-Linear, PET-Surface-Longitudinal, Harmonization & Refactor

Changes

Clinica core:

• [Enh] Code source was completely reformatted using the Black code style.
• [Enh] Documentation for the project is now versioned (versions from 0.3.8 are publicly available).
• [Enh] Functions used for multiple pipelines are now mutualized (e.g container_from_filename function).
• [Enh] f-strings are used massively.

Pipelines:

• [New] pet-linear pipeline: spatial normalization to the MNI space and intensity normalization of PET images. More info in the wiki: https://aramislab.paris.inria.fr/clinica/docs/public/latest/Pipelines/PET_Linear/
• [New] pet-surface-longitudinal pipeline: Surface-based longitudinal processing of PET images [Marcoux et al., 2018]. More info in the wiki: https://aramislab.paris.inria.fr/clinica/docs/public/latest/Pipelines/PET_Surface_Longitudinal/
• [New] Harmonize output message display when running a pipeline.
• [New] Automatically ignore an image to process if it is found in the CAPS folder.
• [Fix] Fix a minor bug when the pet-volume pipeline performs PVC.

Converters:

• [Enh] Conversion information is added once the converter is run to facilitate traceability.
• [Enh] Add new keywords available in ADNI3 to the adni-2-bids converter.

IOtools:

• [New] check-missing-processing tool allows creating a TSV file containing information about the pipelines executed into a specific CAPS folder.