The article discusses the latest advancements in the Clinica software platform, which is designed to automate and streamline the processing and statistical analysis of neuroimaging data. Clinica aims to facilitate the development of machine learning methods for neuroimaging studies. The new features and improvements include: 1. **New Pipelines**: - **fmri-preprocessing**: Corrects fMRI datasets for head motion, slice-timing, and susceptibility distortion, and performs coregistration, spatial normalization, and smoothing. - **dwi-connectome**: Computes a structural connectome using the constrained spherical deconvolution diffusion model, probabilistic tractography, and the Desikan & Destrieux atlases. - **pet-surface**: Analyzes PET data on the cortical surface, including co-registration, intensity normalization, partial volume correction, and spatial normalization. - **machinelearning-prepare-spatial-svm**: Prepares T1 MRI and PET data for classification with an SVM that accounts for spatial and anatomical structure. 2. **Core Upgrades**: - Transitioned to Python 3 and Nipype 1. - Developed a robust continuous integration framework to address challenges in medical image processing, such as reproducibility and stochastic models. 3. **Results**: - The new pipelines expand the scope of modalities supported by Clinica and offer enhanced functionalities. - Clinica has been used in clinical research studies and for developing machine learning frameworks for Alzheimer's disease diagnosis. 4. **Conclusion**: - Clinica is an open-source software platform aimed at making clinical neuroscience studies reproducible and easier to conduct, targeting neuroscientists, clinicians, and machine learning researchers. The article highlights the importance of these advancements in advancing the field of clinical neuroimaging and machine learning.The article discusses the latest advancements in the Clinica software platform, which is designed to automate and streamline the processing and statistical analysis of neuroimaging data. Clinica aims to facilitate the development of machine learning methods for neuroimaging studies. The new features and improvements include: 1. **New Pipelines**: - **fmri-preprocessing**: Corrects fMRI datasets for head motion, slice-timing, and susceptibility distortion, and performs coregistration, spatial normalization, and smoothing. - **dwi-connectome**: Computes a structural connectome using the constrained spherical deconvolution diffusion model, probabilistic tractography, and the Desikan & Destrieux atlases. - **pet-surface**: Analyzes PET data on the cortical surface, including co-registration, intensity normalization, partial volume correction, and spatial normalization. - **machinelearning-prepare-spatial-svm**: Prepares T1 MRI and PET data for classification with an SVM that accounts for spatial and anatomical structure. 2. **Core Upgrades**: - Transitioned to Python 3 and Nipype 1. - Developed a robust continuous integration framework to address challenges in medical image processing, such as reproducibility and stochastic models. 3. **Results**: - The new pipelines expand the scope of modalities supported by Clinica and offer enhanced functionalities. - Clinica has been used in clinical research studies and for developing machine learning frameworks for Alzheimer's disease diagnosis. 4. **Conclusion**: - Clinica is an open-source software platform aimed at making clinical neuroscience studies reproducible and easier to conduct, targeting neuroscientists, clinicians, and machine learning researchers. The article highlights the importance of these advancements in advancing the field of clinical neuroimaging and machine learning.