Funding information
Link to the ANR websiteANR Young Researchers
2019-2024 (48+6 months)
Funding=251k€
Overview
Unsupervised representation learning is suited for knowledge discovery and stratifying risk among medical populations, but faces complex data integration issues. The data descriptors are numerous, high-dimensional and of heterogeneous types, and their combination is not straightforward. MIC-MAC proposes to revisit the data integration approach, by better considering hierarchy (either existing or to be learnt) in the input imaging data. The project is centered on cardiac imaging applications, and plans the retrospective exploration of large existing imaging studies of heart failure patients, from widespread imaging protocols (magnetic resonance [with CHU St Etienne, France] and echocardiography [with Hospital Clínic Barcelona, Spain]).
Project members
- Nicolas DUCHATEAU - Associate Professor (CREATIS) - PI.
- Patrick CLARYSSE - CNRS Research Director (CREATIS).
- Magalie VIALLON - Medical Physicist (CREATIS, CHU St Etienne).
- Pierre CROISILLE - Radiologist, PUPH (CREATIS, CHU St Etienne).
- Benoit FREICHE - PhD student, since 2019.
- Gabriel BERNARDINO - postdoc, 2020-2022.
Publications
Journals:
Hierarchical data integration with Gaussian processes: application to the characterization of cardiac ischemia-reperfusion patterns.
Freiche B, Bernardino G, Deleat-Besson R, Clarysse P, Duchateau N.
IEEE Transactions on Medical Imaging 2024. In press.
Challenges for augmenting intelligence in cardiac imaging.
Sengupta P, Dey D, Davies RH, Duchateau N, Yanamala N.
Lancet Digital Health 2024.
Pixel-wise statistical analysis of myocardial injury in STEMI patients with delayed enhancement MRI.
Duchateau N, Viallon M, Petrusca L, Clarysse P, Mewton N, Belle L, Croisille P.
Frontiers in Cardiovascular Medicine 2023;10.
Characterizing interactions between cardiac shape and deformation by non-linear manifold learning.
Di Folco M, Moceri P, Clarysse P, Duchateau N.
Medical Image Analysis 2022;75:102278.
Additional prognostic value of echocardiographic follow-up in pulmonary hypertension - role of 3D right ventricular area strain.
Moceri P, Duchateau N, Baudouy D, Squara F, Bun SS, Ferrari E, Sermesant M.
European Heart Journal Cardiovascular Imaging 2022;23:1562-72.
Machine learning approaches for myocardial motion and deformation analysis.
Duchateau N, King A, De Craene M.
Frontiers in Cardiovascular Medicine 2020;6:190.
Conference articles:
Domain adaptation of echocardiography segmentation via reinforcement learning.
Judge A, Judge T, Duchateau N, Sandler R, Sokol J, Bernard O, Jodoin PM.
Proc. International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), LNCS 2024;15009:235-44.
Which anatomical directions to quantify local right ventricular strain in 3D echocardiography?
Di Folco M, Dargent T, Bernardino G, Clarysse P, Duchateau N.
Proc. International Conference on Functional Imaging and Modeling of the Heart (FIMH), LNCS 2023;13958:607-15.
Strainger things: discrete differential geometry for transporting right ventricular deformation across surface meshes.
Bernardino G, Dargent T, Camara O, Duchateau N.
Proc. International Conference on Functional Imaging and Modeling of the Heart (FIMH), LNCS 2023;13958:338-46
Assessment of the evolution of temporal segmental strain in a longitudinal study of myocardial infarction.
Freytag B, Duchateau N, Petrusca L, Ohayon J, Croisille P, Clarysse P.
Proc. International Conference on Functional Imaging and Modeling of the Heart (FIMH), LNCS 2023;13958:678-87.
Localizing cardiac dyssynchrony in M-mode echocardiography with attention maps.
Saiz-Vivó M, Capallera I, Duchateau N, Bernardino G, Piella G, Camara O.
Proc. International Conference on Functional Imaging and Modeling of the Heart (FIMH), LNCS 2023;13958:688-97.
Reinforcement learning for active modality selection during diagnosis.
Bernardino G, Jonsson A, Loncaric F, Martí Castellote PM, Sitges M, Clarysse P, Duchateau N.
Proc. International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), LNCS 2022;13431:592-601.
Characterizing myocardial ischemia and reperfusion patterns with hierarchical manifold learning.
Freiche B, Clarysse P, Viallon M, Croisille P, Duchateau N.
Proc. Statistical Atlases and Computational Models of the Heart (STACOM), MICCAI’21 Workshop, LNCS 2022;13131:66-74.
Hierarchical multi-modality prediction model to assess obesity-related remodelling.
Bernardino G, Clarysse P, Sepúlveda-Martı́nez A, Rodrı́guez-López M, Prat-Gonzàlez S, Sitges M, Gratacós E, Crispi F, Duchateau N.
Proc. Statistical Atlases and Computational Models of the Heart (STACOM), MICCAI’21 Workshop, LNCS 2022;13131:103-12.
Population-based personalization of geometric models of myocardial infarction.
Mom K, Clarysse P, Duchateau N.
Proc. International Conference on Functional Imaging and Modeling of the Heart (FIMH), LNCS 2021;12738:3-11.
Investigation of the impact of normalization on the study of interactions between myocardial shape and deformation.
Di Folco M, Guigui N, Clarysse P, Moceri P, Duchateau N.
Proc. International Conference on Functional Imaging and Modeling of the Heart (FIMH), LNCS 2021;12738:223-31.
Abstracts:
Adaptation de domaine de segmentations échocardiographiques via l’apprentissage par renforcement
Judge A, Judge T, Duchateau N, Sandler R, Sokol J, Bernard O, Jodoin PM.
Journée du GDR IASIS “Vers un apprentissage pragmatique dans un contexte de données visuelles labellisées limitées”, 2024.
Fusing echocardiography images and medical records for continuous patient stratification
Painchaud N, Courand PY, Jodoin PM, Duchateau N, Bernard O.
Medical Imaging with Deep Learning (MIDL) - short papers, 2024.
Apprentissage de représentation pour la quantification de biais dans l’analyse de populations: application à la comparaison du LGE conventionnel vs. synthétique avec TI optimal.
Deleat-Besson R, Viallon M, Petrusca L, Croisille P, Duchateau N.
Colloque Français d'Intelligence Artificielle en Imagerie Biomédicale (IABM) 2024.
Fusion d'échocardiographie et de dossiers médicaux pour la caractérisation de l'hypertension.
Painchaud N, Courand PY, Jodoin PM, Duchateau N, Bernard O.
Colloque Français d'Intelligence Artificielle en Imagerie Biomédicale (IABM) 2024.
AI-based comparison of conventional LGE & synthetic MagIR-LGE with optimal inversion-time: impact on population analysis?
Deleat-Besson R, Viallon M, Petrusca L, Croisille P, Duchateau N.
Society for Cardiovascular Magnetic Resonance (SCMR) congress 2023.
Machine learning for the generation of personalized image analysis protocol in echocardiography - a pilot study in arterial hypertension.
Bernardino G , Loncaric F, Jonsson A, Castellote PM, Sitges M, Clarysse P, Duchateau N.
European Heart Journal: Cardiovascular Imaging, Abstracts from the EuroEcho Congress 2023;24:i558-9.
Hierarchical manifold learning for the interpretation of multi-level data - Application to cardiac imaging.
Freiche B, Clarysse P, Viallon M, Croisille P, Duchateau N.
Medical Image Analysis and Artificial Intelligence (MAI), Sino-French workshop 2021.
Pixel-wise statistical analysis of lesion patterns: a fresh look at immediate vs. delayed stenting of the Minimalist Immediate Mechanical Intervention approach (MIMI) in acute STEMI.
Duchateau N, Viallon M, Petrusca L, Clarysse P, Belle L, Croisille P.
Society for Cardiovascular Magnetic Resonance (SCMR) congress 2021.
Books:
Functional Imaging and Modeling of the Heart (FIMH'23).
Bernard O, Clarysse P, Duchateau N, Ohayon J, Viallon M, eds.
Springer, LNCS, 2023;13958.
AI and Big Data in cardiology: a practical guide.
Duchateau N, King A, eds.
Springer, 2023. In press.
Including the following chapters:
Conclusion.
King A, Duchateau N.
Analysis of non-imaging data.
Duchateau N, Camara O, Sebastian R, King A.
Outcome prediction.
Ly B, Pop M, Cochet H, Duchateau N, O’Regan D, Sermesant M.
Diagnosis.
Rueckert D, Knolle M, Duchateau N, Razavi R, Kaissis G.
From machine learning to deep learning.
Jodoin PM, Duchateau N, Desrosiers C.
AI and machine learning: the basics.
Duchateau N, Puyol-Antón E, Ruijsink B, King A.
Introduction.
King A, Duchateau N.
Other:
Machine learning and biophysical models: how to benefit each other?
Duchateau N, Camara O.
In: Chinesta F, Cueto E, Payan Y, Ohayon J, eds. Reduced order models for the biomechanics of living organs. Elsevier, 2023:147-64
Contact Us
CREATIS, INSA Bâtiment Léonard de Vinci - 21, Avenue J. Capelle - 69621 Villeurbanne Cedex, France.
Phone: +33.472437147
Email: nicolas.duchateau [at] creatis.insa-lyon.fr