There are many active research projects accessing and applying shared ADNI data. Use the search above to find specific research focuses on the active ADNI investigations. This information is requested annually as a requirement for data access.
Principal Investigator | |
Principal Investigator's Name: | Alexandre Routier |
Institution: | Paris Brain Institute / Inria |
Department: | Aramis Lab |
Country: | |
Proposed Analysis: | We would like to use the ADNI dataset as a support for educational courses for the reproducible classification of Alzheimer's disease using machine learning or deep learning approaches. This topic was evaluated in previous papers by our team (Aramis Lab, Paris Brain Institute) on machine learning (Samper J. et al., NeuroImage, 2018; https://doi.org/10.1016/j.neuroimage.2018.08.042) and deep learning (Wen J., Thibeau-Sutre E. et al., Medical Image Analysis, 2020; https://arxiv.org/abs/1904.07773) The target audience would be attendees of educational courses at conferences (e.g. MICCAI, OHBM) or Master students following IA courses on medical imaging. On a short term basis, we would like to participate to the MICCAI Educational Challenge (https://miccai-sb.github.io/challenge.html) where an article and/or a notebook would explain what image pre-processing steps should be performed, how to design a cross validation and avoid data leakage, how to make a rigorous comparison of different classification methods, all using the Clinica software (http://www.clinica.run/). Clinica is a software platform for clinical research studies involving patients with neurological and psychiatric diseases and the acquisition of multimodal data. Clinica provides a converter of ADNI dataset into the BIDS format. It also provides a set of pipelines to process neuroimaging data (T1-weighted MRI, diffusion MRI, PET data) based on software tools developed in the neuroimaging community (ANTs, FreeSurfer, FSL, MRtrix, PETPVC, SPM). It also provides integration between feature extraction and statistics, machine learning or deep learning. Clinica is showcased as a framework for the reproducible classification of Alzheimer's disease using machine learning (https://github.com/aramis-lab/AD-ML developed for (Samper J. et al., 2018)) and deep learning (https://github.com/aramis-lab/AD-DL developed for (Wen J., Thibeau-Sutre E. et al., 2020)). We already organized courses covering this topic, such as a workshop on machine learning applied to medical imaging organized in March 2020 (https://laclauc.github.io/workshop.html ). To build these educational courses, we would like to redistribute the following features extracted by Clinica: Gray matter maps (segmented from T1-weighted MRI) PyTorch tensor versions of these segmentations TSV files containing mean regional measures of different brain parcellations (e.g. mean gray matter maps, mean SUVR, mean DTI-based measures) We would also need to provide demographic, diagnostic and clinical data. If these data are considered as sensitive, we will anonymize ADNI identifiers before distribution. Data used for these courses would be available through a server hosted by Inria, the French National Institute for Research in Digital Science and Technology (www.inria.fr/en). The download of the data will be restricted to course participants. |
Additional Investigators | |
Investigator's Name: | Olivier Colliot |
Proposed Analysis: | We would like to use the ADNI dataset as a support for educational courses for the reproducible classification of Alzheimer's disease using machine learning or deep learning approaches. This topic was evaluated in previous papers by our team (Aramis Lab, Paris Brain Institute) on machine learning (Samper J. et al., NeuroImage, 2018; https://doi.org/10.1016/j.neuroimage.2018.08.042) and deep learning (Wen J., Thibeau-Sutre E. et al., Medical Image Analysis, 2020; https://arxiv.org/abs/1904.07773) The target audience would be attendees of educational courses at conferences (e.g. MICCAI, OHBM) or Master students following IA courses on medical imaging. On a short term basis, we would like to participate to the MICCAI Educational Challenge (https://miccai-sb.github.io/challenge.html) where an article and/or a notebook would explain what image pre-processing steps should be performed, how to design a cross validation and avoid data leakage, how to make a rigorous comparison of different classification methods, all using the Clinica software (http://www.clinica.run/). Clinica is a software platform for clinical research studies involving patients with neurological and psychiatric diseases and the acquisition of multimodal data. Clinica provides a converter of ADNI dataset into the BIDS format. It also provides a set of pipelines to process neuroimaging data (T1-weighted MRI, diffusion MRI, PET data) based on software tools developed in the neuroimaging community (ANTs, FreeSurfer, FSL, MRtrix, PETPVC, SPM). It also provides integration between feature extraction and statistics, machine learning or deep learning. Clinica is showcased as a framework for the reproducible classification of Alzheimer's disease using machine learning (https://github.com/aramis-lab/AD-ML developed for (Samper J. et al., 2018)) and deep learning (https://github.com/aramis-lab/AD-DL developed for (Wen J., Thibeau-Sutre E. et al., 2020)). We already organized courses covering this topic, such as a workshop on machine learning applied to medical imaging organized in March 2020 (https://laclauc.github.io/workshop.html ). To build these educational courses, we would like to redistribute the following features extracted by Clinica: Gray matter maps (segmented from T1-weighted MRI) PyTorch tensor versions of these segmentations TSV files containing mean regional measures of different brain parcellations (e.g. mean gray matter maps, mean SUVR, mean DTI-based measures) We would also need to provide demographic, diagnostic and clinical data. If these data are considered as sensitive, we will anonymize ADNI identifiers before distribution. Data used for these courses would be available through a server hosted by Inria, the French National Institute for Research in Digital Science and Technology (www.inria.fr/en). The download of the data will be restricted to course participants. |
Investigator's Name: | Ninon Burgos |
Proposed Analysis: | We would like to use the ADNI dataset as a support for educational courses for the reproducible classification of Alzheimer's disease using machine learning or deep learning approaches. This topic was evaluated in previous papers by our team (Aramis Lab, Paris Brain Institute) on machine learning (Samper J. et al., NeuroImage, 2018; https://doi.org/10.1016/j.neuroimage.2018.08.042) and deep learning (Wen J., Thibeau-Sutre E. et al., Medical Image Analysis, 2020; https://arxiv.org/abs/1904.07773) The target audience would be attendees of educational courses at conferences (e.g. MICCAI, OHBM) or Master students following IA courses on medical imaging. On a short term basis, we would like to participate to the MICCAI Educational Challenge (https://miccai-sb.github.io/challenge.html) where an article and/or a notebook would explain what image pre-processing steps should be performed, how to design a cross validation and avoid data leakage, how to make a rigorous comparison of different classification methods, all using the Clinica software (http://www.clinica.run/). Clinica is a software platform for clinical research studies involving patients with neurological and psychiatric diseases and the acquisition of multimodal data. Clinica provides a converter of ADNI dataset into the BIDS format. It also provides a set of pipelines to process neuroimaging data (T1-weighted MRI, diffusion MRI, PET data) based on software tools developed in the neuroimaging community (ANTs, FreeSurfer, FSL, MRtrix, PETPVC, SPM). It also provides integration between feature extraction and statistics, machine learning or deep learning. Clinica is showcased as a framework for the reproducible classification of Alzheimer's disease using machine learning (https://github.com/aramis-lab/AD-ML developed for (Samper J. et al., 2018)) and deep learning (https://github.com/aramis-lab/AD-DL developed for (Wen J., Thibeau-Sutre E. et al., 2020)). We already organized courses covering this topic, such as a workshop on machine learning applied to medical imaging organized in March 2020 (https://laclauc.github.io/workshop.html ). To build these educational courses, we would like to redistribute the following features extracted by Clinica: Gray matter maps (segmented from T1-weighted MRI) PyTorch tensor versions of these segmentations TSV files containing mean regional measures of different brain parcellations (e.g. mean gray matter maps, mean SUVR, mean DTI-based measures) We would also need to provide demographic, diagnostic and clinical data. If these data are considered as sensitive, we will anonymize ADNI identifiers before distribution. Data used for these courses would be available through a server hosted by Inria, the French National Institute for Research in Digital Science and Technology (www.inria.fr/en). The download of the data will be restricted to course participants. |