Sophie Giffard-Roisin

IRD researcher

Contact

sophie.giffard['at']univ-grenoble-alpes.fr

Scholar account

Github account

Gitlab UGA account

ISTerre, Grenoble Universités

Bureau 223 (0476635233)

1381 Rue de la Piscine

38610 Gières, France

Hello!

I joined ISTerre Laboratory, Grenoble, France in October 2019 as an IRD (Institut de Recherche pour le Développement) researcher!

Research Interests. My research focuses on applied maths and computer science, and more particularly machine learning techniques for natural hazards, solid earth and climate informatics. My previous research also included medical image analysis, machine learning for cardiac personalisation and inverse problems.

Looking for a post-doc, PhD or M2 internship? If you want to work on machine learning for natural hazards applications, using satellite images, seismic and/or GPS signals, contact me. I am often looking for motivated students, either computer scientists wanting to work on impactful problems, or nerdy geoscientists!

The ANR JCJC Project EDAM: Earth deformation from automatic mapping, will start on Oct. 2024! We will work for the next 4 years on the development of specific machine learning models to detect at a large scale geomorphological objects (passed faults and landslides fossils), as well as currently moving objects (slow moving landslides, earthquakes) by fusing different satellite imagery. Our targetted region: South Peru.

2-year Post-doc offer for 2025: check here

I will be a visiting researcher at Columbia University in the Lamont Observatory between Feb. 2025 and Feb. 2026 thanks to a Fulbright Fellowship.

Bio

• 2019- .. IRD chargée de recherche position, ISTerre laboratory, Université Grenoble Alpes.
• 2018-2019. Post-doctoral position on climate machine learning at Colorado University, Boulder, USA and LAL, Orsay Paris-Sud (with Claire Monteleoni, Balázs Kégl and Guillaume Charpiat)
• 2014-2017. PhD candidate (PhD defense: Dec. 2017) on non-invasive personalisation of cardiac model at Inria Sophia-Antipolis (France) in the Asclepios/Epione team (under the supervision of M. Sermesant, H. Delingette and N. Ayache).

My full resume is available here. (might not be up-to-date)

Publications

Pre-prints:

• Automatic characterization of normal fault scarps using convolutional neural networks
  Lea Pousse-Beltran, Théo Lallemand, Laurence Audin, Pierre Lacan, Andres David Nuñez-Meneses, Sophie Giffard-Roisin. under review pdf link
• Fast And Accurate Sub-pixel Displacement Estimation From Optical Satellite Images Using A New Hyper-realistic Earthquake Database And U-net Architecture
  Tristan Montagnon, James Hollingsworth, Erwan Pathier, Mathilde Marchandon, Mauro Dalla Mura, Sophie Giffard-Roisin. IGARSS 2024 pdf link, poster download
• Sub-Pixel Displacement Estimation with Deep Learning: Application to Optical Satellite Images Containing Sharp Displacements
  Tristan Montagnon, Sophie Giffard-Roisin, Mauro Dalla Mura, Mathilde Marchandon, Erwan Pathier, James Hollingsworth. under revision, 2024 pdf link
• Denoising of Geodetic Time Series Using Spatiotemporal Graph Neural Networks: Application to Slow Slip Event Extraction
  Giuseppe Costantino, Sophie Giffard-Roisin, Mauro Dalla Mura, Anne Socquet. under revision, 2024 arXiv preprint

Journal publications:

• Deep learning detects uncataloged low-frequency earthquakes across regions
  Jannes Münchmeyer, Sophie Giffard-Roisin, Marielle Malfante, William Frank, Piero Poli, David Marsan, Anne Socquet. Seismica 2024 paper link
• Multi-station deep learning on geodetic time series detects slow slip events in Cascadia
  Giuseppe Costantino, Sophie Giffard-Roisin, Mathilde Radiguet, Mauro Dalla Mura, David Marsan, Anne Socquet. Nature Communications Earth & Environment, vol. 4, p 435, 2023. open-access link
• Seismic source characterization from GNSS data using deep learning
  Giuseppe Costantino, Sophie Giffard-Roisin, David Marsan, Mathilde Radiguet, Mauro Dalla Mura, Anne Socquet. Journal of Geophysical Research - Solid Earth, 2023. open-access link
• Classification of red cell dynamics with convolutional and recurrent neural networks: a sickle cell disease case study
  Maxime Darrin, Ashwin Samudre, Maxime Sahun, Scott Atwell, Catherine Badens, Anne Charrier, Emmanuèle Helfer, Annie Viallat, Vincent Cohen-Addad, Sophie Giffard-Roisin. Scientific Reports, 13(1), 745, 2023. open-access link
• Interpreting convolutional neural network decision for earthquake detection with feature map visualisation, backward optimisation and layer-wise relevance propagation methods
  Josipa Majstorović, Sophie Giffard-Roisin, Piero Poli. Geophysical Journal International, 2022. pdf link
• Land cover classification of the Alps from InSAR temporal coherence matrices
  Sophie Giffard-Roisin, SalahEddine Boudaour, Marie-Pierre Doin, Yajing Yan, Abdourrahmane Atto. Frontiers in Remote Sensing, 2022. pdf link
• Testing machine learning models for seismic damage prediction at a regional scale using building-damage dataset compiled after the 2015 Gorkha Nepal earthquake
  Subash Ghimire, Philippe Guéguen, Sophie Giffard-Roisin, and Danijel Schorlemmer Earthquake Spectra, 2022. pdf link
• Designing convolutional neural network pipeline for near-fault earthquake catalog extension using single-station waveforms.
  Josipa Majstorović, Sophie Giffard-Roisin, Piero Poli. Journal of Geophysical Research: Solid Earth, 2021. https://doi.org/10.1029/2020JB021566
• Automatic Color Detection-Based Method Applied to Sentinel-1 SAR Images for Snow Avalanche Debris Monitoring.
  Anna Karas, Fatima Karbou, Sophie Giffard-Roisin, Philippe Durand, Nicolas Eckert. IEEE Transactions on Geoscience and Remote Sensing, 2021. https://doi.org/10.1109/TGRS.2021.3131853
• Tropical Cyclone Track Forecasting using Fused Deep Learning from Aligned Reanalysis Data.
  Sophie Giffard-Roisin, Mo Yang, Guillaume Charpiat, Christina Kumler-Bonfanti, Balázs Kégl, Claire Monteleoni. Frontiers in Big Data - Data-driven Climate Sciences, 2020. arXiv link
• Transfer Learning from Simulations on a Reference Anatomy for ECGI in Personalised Cardiac Resynchronization Therapy.
  Sophie Giffard-Roisin, Hervé Delingette, Thomas Jackson, Jessica Webb, Lauren Fovargue, Jack Lee, Christopher A. Rinaldi, Reza Razavi, Nicholas Ayache, Maxime Sermesant. IEEE Transactions on Biomedical Engineering, 2019. HAL link
• A Rule-based Method to Model Myocardial Fiber Orientation in Cardiac Biventricular Geometries with Outflow Tracts.
  Ruben Doste, David Soto-Iglesias, Gabriel Bernardino, Alejandro Alcaine, Rafael Sebastian, Sophie Giffard-Roisin, Maxime Sermesant, Antonio Berruezo, Damian Sanchez-Quintana, Oscar Camara. International Journal for Numerical Methods in Biomedical Engineering, 2019. Wiley link
• Noninvasive Personalization of a Cardiac Electrophysiology Model From Body Surface Potential Mapping.
  Sophie Giffard-Roisin, Thomas Jackson, Lauren Fovargue, Jack Lee, Hervé Delingette, Reza Razavi, Nicholas Ayache, and Maxime Sermesant. IEEE Transactions on Biomedical Engineering, 2017. HAL link
• A Framework for the Generation of Realistic Synthetic Cardiac Ultrasound and Magnetic Resonance Imaging Sequences from the same Virtual Patients.
  Yitian Zhou, Sophie Giffard-Roisin, Mathieu De Craene, Jan D’hooge, Martino Alessandrini, Denis Friboulet, Maxime Sermesant and Olivier Bernard. IEEE Transactions on Medical Imaging, 2017. HAL link
• A Pipeline for the Generation of Realistic 3D Synthetic Echocardiographic Sequences: Methodology and Open-Access Database.
  Martino Alessandrini, Mathieu De Craene, Olivier Bernard, Sophie Giffard-Roisin, Pascal Allain, Irina Waechter-Stehle, Jürgen Weese, Eric Saloux, Hervé Delingette, Maxime Sermesant and Jan D’Hooge. IEEE Transactions on Medical Imaging, 2015. HAL link

Proceedings (conferences & workshops):

• A new deep-learning approach for the sub-pixel registration of satellite images containing sharp displacement discontinuities
  Tristan Montagnon, James Hollingsworth, Erwan Pathier, Mathilde Marchandon, Mauro Dalla Mura, Sophie Giffard-Roisin. IGARSS 2023-2023 IEEE International Geoscience and Remote Sensing Symposium, 2023/7/16 pdf link
• Characterization of Slow Slip Events from Gnss Data with Deep Learning
  Giuseppe Costantino, Sophie Giffard-Roisin, Mauro Dalla Mura, Anne Socquet. IGARSS 2023-2023 IEEE International Geoscience and Remote Sensing Symposium, 2023/7/16 ieee link
• Characterization of normal fault scarp using convolutional neural network: application to Mexico
  Lea Pousse-Beltran, Sophie Giffard-Roisin, Laurence Audin, Pierre Lacan, Théo Lallemand, Glenn Cougoulat, Andrés Núñez Meneses. 11th International INQUA Meeting on Paleoseismology, Active Tectonics and Archeoseismology (PATA), INQUA Focus Group Terrestrial Processes Perturbed by Tectonics (TPPT), Sept. 2022. pdf link
• Optical Satellite Image Registration for Ground Deformation using Deep Learning
  Tristan Montagnon, James Hollingsworth, Erwan Pathier, Mathilde Marchandon, Mauro Dalla Mura, Sophie Giffard-Roisin. ICIP IEEE International Conference on Image Processing, Oct. 2022 pdf link
• Variational Autoencoder Anomaly-Detection of Avalanche Deposits in Satellite SAR Imagery.
  Saumya Sinha, Sophie Giffard-Roisin, Fatima Karbou, Michaël Deschatres, Anna Karas, Nicolas Eckert, Claire Monteleoni. CI2020: Proceedings of the 10th International Conference on Climate Informatics, Sep. 2020. https://doi.org/10.1145/3429309.3429326 ACM link
• Detecting Avalanche Deposits using Variational Autoencoder on Sentinel-1 Satellite Imagery.
  Saumya Sinha, Sophie Giffard-Roisin, Fatima Karbou, Michaël Deschatres, Anna Karas, Nicolas Eckert, Claire Monteleoni. Tackling Climate Change with Machine Learning NeurIPS workshop, Vancouver, Dec. 2019. *Spotlight talk selection!* HAL link
• Can Avalanche Deposits be Effectively Detected by Deep Learning on Sentinel-1 Satellite SAR Images?
  Saumya Sinha*, Sophie Giffard-Roisin*, Fatima Karbou, Michael Deschatres, Anna Karas, Nicolas Eckert, Cécile Coléou, Claire Monteleoni. Climate Informatics Workshop Proceedings, Paris, Oct 2019. HAL link
• Deep Learning for Hurricane Track Forecasting from Aligned Spatio-temporal Climate Datasets.
  Sophie Giffard-Roisin*, Mo Yang*, Guillaume Charpiat, Balázs Kégl, Claire Monteleoni. Modeling and decision-making in the spatiotemporal domain NIPS workshop proceedings, Montreal, Dec. 2018. HAL link Codes: Github link
• The 2018 Climate Informatics Hackathon: Hurricane Intensity Forecast.
  Sophie Giffard-Roisin, David Gagne, Alexandre Boucaud, Balázs Kégl, Mo Yang, Guillaume Charpiat and Claire Monteleoni. Climate Informatics Workshop Proceedings, Boulder, Sept. 2018. HAL link
• Fused Deep Learning for Hurricane Track Forecast from Reanalysis Data.
  Sophie Giffard-Roisin*, Mo Yang*, Guillaume Charpiat, Balázs Kégl, Claire Monteleoni. Climate Informatics Workshop Proceedings, Boulder, Sept. 2018. HAL link
• Estimation of the Spatial Resolution of a 2D Strain Estimator Using Synthetic Cardiac Images.
  Bidisha Chakraborty, Sophie Giffard-Roisin, Martino Alessandrini, Brecht Heyde, Maxime Sermesant, Jan D’hooge. In 2018 IEEE International Ultrasonics Symposium (IUS) Proceedings, Kobe, Japan, 2018. HAL link
• Sparse Bayesian Non-linear Regression for Multiple Onsets Estimation in Non-invasive Cardiac Electrophysiology.
  Sophie Giffard-Roisin, Hervé Delingette, Thomas Jackson, Lauren Fovargue, Jack Lee, Aldo Rinaldi, Nicholas Ayache, Reza Razavi, and Maxime Sermesant. In Functional imaging and modelling of the heart (FIMH), 2017. Best paper award! HAL link
• Smoothed Particle Hydrodynamics for Electrophysiological Modeling: an Alternative to Finite Element Methods.
  Èric Lluch Alvarez, Rubén Doste, Sophie Giffard-Roisin, Alexandre This, Maxime Sermesant, Oscar Camara, Mathieu De Craene and Hernán G. Morales. Functional imaging and modelling of the heart (FIMH) conference, 2017. HAL link
• A Rule-Based Method to Model Myocardial Fiber Orientation for Simulating Ventricular Out-flow Tract Arrhythmias.
  Rubén Doste, David Soto-Iglesias, Gabriel Bernardino, Rafael Sebastian, Sophie Giffard-Roisin, Rocio Cabrera-Lozoya, Maxime Sermesant, Antonio Berruezo, Damián Sánchez-Quintana and Oscar Camara, Functional imaging and modelling of the heart (FIMH) conference, 2017. HAL link
• Estimation of Purkinje Activation from ECG: an Intermittent Left Bundle Branch Block Study.
  Sophie Giffard-Roisin, Lauren Fovargue, Jessica Webb, Roch Molléro, Jack Lee, Hervé Delingette, Nicholas Ayache, Reza Razavi, and Maxime Sermesant. STACOM proceedings, held in conjunction with MICCAI, Athens, 2016. HAL link
• Novel Framework to Integrate Real-Time MR-Guided EP Data with T1 Mapping-Based Computational Heart Models.
  Sebastian Ferguson, Maxime Sermesant, Samuel Oduneye, Sophie Giffard-Roisin, Michael Truong, Labonny Biswas, Nicholas Ayache, Graham Wright, Mihaela Pop. STACOM proceedings, Held in Conjunction with MICCAI, Athens, 2016. SemanticScholar link
• Generation of Ultra-realistic Synthetic Echocardiographic Sequences to Facilitate Standardization of Deformation Imaging.
  Martino Alessandrini, and Brecht Heyde, Sophie Giffard-Roisin, Hervé Delingette, Maxime Sermesant, Pascal Allain, Olivier Bernard, Mathieu De Craene, Jan D’hooge, 12th International Symposium on Biomedical Imaging (ISBI), 2015. HAL link
• Evaluation of Personalised Canine Electromechanical Models.
  Sophie Giffard-Roisin, Stéphanie Marchesseau, Loic Le Folgoc, Hervé Delingette, and Maxime Sermesant. STACOM proceedings, held in conjunction with MICCAI, Boston, 2014. HAL link

Abstacts and Posters:

• Towards the characterization of Slow Slip deformation by means of deep learning.
  Giuseppe Costantino, Sophie Giffard-Roisin, Mauro Dalla Mura, David Marsan, Mathilde Radiguet, Anne Socquet, EGU General Assembly 2022. Abstract link
• Analysis of the potential correlation between intraslab intermediate-depth and shallow earthquakes in the Japan trench subduction zone prior to the Mw 9.0 Tohoku-oki earthquake
  Audrey Chouli, David Marsan, Sophie Giffard-Roisin, Michel Bouchon, Anne Socquet. EGU General Assembly 2022. Abstract link
• Revealing Glacier Dynamics by Hierarchical Clustering of Continuous Seismic Data Recorded on a Dense Seismic Array: Application to Argentiere Glacier, French Alps.
  V Shalaeva, P Poli, S Giffard-Roisin, S Garambois. AGU Fall Meeting Abstracts 2021, S35C-0233 Abstract link
• Towards assessing the link between slow slip and seismicity with a Deep Learning approach.
  Giuseppe Costantino, Mauro Dalla Mura, David Marsan, Sophie Giffard-Roisin, Mathilde Radiguet, and Anne Socquet, EGU General Assembly 2020. Abstract link
• Detecting avalanche debris from SAR imaging: a comparison of convolutional neural networks and variational autoencoders.
  Sophie Giffard-Roisin, Saumya Sinha, Fatima Karbou, Michael Deschatres, Anna Karas, Nicolas Eckert, Cécile Coléou, and Claire Monteleoni, EGU General Assembly 2020. Abstract link
• Learning to automatically detect avalanche deposition from SAR satellite imagery.
  Sophie Giffard-Roisin*, Saumya Sinha*, Nicolas Eckert, Michael Dechartres, Cécile Coléou, Claire Monteleoni, Fatima Karbou, NOAA Workshop on Leveraging AI in the Exploitation of Satellite Earth Observations & Numerical Weather Prediction. Washington DC, 2019. Poster link
• Noninvasive localization of premature ventricular complexes: a research-community-based approach.
  MJM Cluitmans, S Ghimire, J Dhamala, J Coll-Font, JD Tate, S Giffard-Roisin, J Svehlikova, O Doessel, MS Guillem, DH Brooks, RS Macleod, L Wang, EP Europace 20.suppl 1, 2018. (2 pages) Poster link
• Predicting Acute Cardiac Resynchronisation Therapy Effects through Patient Specific Modelling.
  Lauren Fovargue, Simone Rivolo, Jessica Webb, Sophie Giffard-Roisin, Simon Clairidge, Tiffany Patterson, Liya Asner, Thomas Jackson, Eric Kerfoot, David Nordsletten, Maxime Sermesant, Reza Razavi, Nicolas P. Smith, Jack Lee, ECCOMAS Congress 2016. (1 page) Abstract link

PhD thesis:

• Non-invasive Personalisation of Cardiac Electrophysiological Models from Surface Electrograms.
  Sophie Giffard-Roisin. UCA - Université Côte d'Azur, 2017. HAL link