New advances in the Clinica software platform for clinical neuroimaging studies. Alexandre Routier, Arnaud Marcoux, Mauricio Diaz Melo, Jérémy Guillon, Jorge Samper-González, Junhao Wen, Simona Bottani, Alexis Guyot, Elina Thibeau-Sutre, Marc Teichmann, et al. The Clinica platform aims to automate the processing and statistical analysis of neuroimaging data and ease the development of machine learning approaches. The paper presents new functionalities and improvements made to Clinica, including framework upgrades, continuous integration and new pipelines (structural connectome, surface-based PET analysis, fMRI preprocessing, machine learning). Clinica is based on the Nipype library and relies on tools from the neuroimaging community to build pipelines (FreeSurfer, FSL, SPM, ANTs, MRtrix3, and PETPVC) or build machine learning modules (scikit-learn). The only assumption on the raw data is that they follow the BIDS format. New functionalities of Clinica can be divided into three main parts: new pipelines, framework upgrades, and continuous integration. New pipelines include fmri-preprocessing, dwi-connectome, pet-surface, and machinelearning-prepare-spatial-svm. The pipelines providing image-derived measurements are listed in Fig 1. The core of Clinica was consolidated and upgraded to use Python 3 and Nipype 1. A robust continuous integration framework was developed, addressing specific challenges raised by medical image processing. The new pipelines expand the scope of modalities covered by Clinica and offer new functionalities. Fig 2 illustrates how pipelines can be chained to carry out a study, from the conversion of the raw imaging data to the statistical analysis. Clinica has already been used for clinical research studies and to develop machine learning frameworks for Alzheimer's disease diagnosis. This is in line with the target audience of Clinica, namely neuroscientists and clinicians conducting neuroimaging studies, and researchers in machine learning. Clinica is an open source software platform for reproducible clinical neuroscience studies. It aims to make clinical research studies easier and pursues the community effort of reproducibility.New advances in the Clinica software platform for clinical neuroimaging studies. Alexandre Routier, Arnaud Marcoux, Mauricio Diaz Melo, Jérémy Guillon, Jorge Samper-González, Junhao Wen, Simona Bottani, Alexis Guyot, Elina Thibeau-Sutre, Marc Teichmann, et al. The Clinica platform aims to automate the processing and statistical analysis of neuroimaging data and ease the development of machine learning approaches. The paper presents new functionalities and improvements made to Clinica, including framework upgrades, continuous integration and new pipelines (structural connectome, surface-based PET analysis, fMRI preprocessing, machine learning). Clinica is based on the Nipype library and relies on tools from the neuroimaging community to build pipelines (FreeSurfer, FSL, SPM, ANTs, MRtrix3, and PETPVC) or build machine learning modules (scikit-learn). The only assumption on the raw data is that they follow the BIDS format. New functionalities of Clinica can be divided into three main parts: new pipelines, framework upgrades, and continuous integration. New pipelines include fmri-preprocessing, dwi-connectome, pet-surface, and machinelearning-prepare-spatial-svm. The pipelines providing image-derived measurements are listed in Fig 1. The core of Clinica was consolidated and upgraded to use Python 3 and Nipype 1. A robust continuous integration framework was developed, addressing specific challenges raised by medical image processing. The new pipelines expand the scope of modalities covered by Clinica and offer new functionalities. Fig 2 illustrates how pipelines can be chained to carry out a study, from the conversion of the raw imaging data to the statistical analysis. Clinica has already been used for clinical research studies and to develop machine learning frameworks for Alzheimer's disease diagnosis. This is in line with the target audience of Clinica, namely neuroscientists and clinicians conducting neuroimaging studies, and researchers in machine learning. Clinica is an open source software platform for reproducible clinical neuroscience studies. It aims to make clinical research studies easier and pursues the community effort of reproducibility.