Publications

Machine Learning

Towards reliable machine learning under domain shift and costly labeling, with applications to engineering design

In engineering design, using machine learning models to find innovative products poses major challenges. The effectiveness of machine learning models has been demonstrated when trained and used on large datasets of independently identically…

Antoine de Mathelin de Papigny

Signal and Image processing

METHANE EMISSIONS MONITORING USING GEOSTATIONARY SATELLITES

Satellite imaging has proven to be crucial to monitor methane emissions and help reduce them. In this paper, we propose an automatic practical methodology to use time series from geostationary satellites like GOES-16. While these satellites offer a…

A. Groshenry, Clément Giron, C. Hessel, C. de Franchis, Gabriele Facciolo, T. Ehret

2024 IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2024) - 2024

Environmental Engineering

MODEL ADJUSTED MATCHED FILTER FOR METHANE PLUME DETECTION ON PRISMA HYPERSPECTRAL IMAGES

Reducing methane emissions is essential to tackle climate change. Here, we address the problem of detecting automatically point source methane leaks using high resolution hyperspectral images from the PRISMA satellite. We propose an improvement of…

E. Ouerghi, T. Ehret, Gabriele Facciolo, E. Meinhardt, C. de Franchis, A. Groshenry, J.-M. Morel

2024 IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2024) - 2024

Analysis of PDEs

Linear elliptic homogenization for a class of highly oscillating non-periodic potentials

We consider an homogenization problem for the second order elliptic equation $- \Delta u^{\varepsilon} + \dfrac{1}{\varepsilon} V(./\varepsilon) u^{\varepsilon} + \nu u^{\varepsilon} =f$ when the highly oscillatory potential $V$ belongs to a…

Rémi Goudey, Claude Le Bris

SIAM Journal on Mathematical Analysis - 2024

Life Sciences

“Hemispheric pilocytic astrocytoma” revisited: A comprehensive clinicopathological and molecular series emphasizing their overlap with other glioneuronal tumors

Abstract Pilocytic astrocytomas (PA) typically exhibit distinct clinical, radiological, histopathological, and genetic features. DNA-methylation profiling distinguishes PA according to their location (infratentorial, midline, hemispheric, or spinal)…

Cassandra Mariet, Jacques Grill, Yassine Ajlil, David Castel, Volodia Dangouloff-Ros, Nathalie Boddaert, Alexandra Meurgey, Daniel Pissaloux, Romain Appay, Raphaël Saffroy, Stéphanie Puget, Thomas Blauwblomme, Kévin Beccaria, Lauren Hasty, Valérie Rigau, Thomas Roujeau, Aude Aline-Fardin, Fabrice Chrétien, Alice Métais, Pascale Varlet, Arnault Tauziède-Espariat

Journal of Neuropathology and Experimental Neurology - 2024

Signal and Image processing

Topological data analysis for time series

In this thesis, we develop new methods for time series analysis based on topological data analysis. This work is motivated by the study of physiological signals. These signals usually have a certain structure that can be studied using tools such as…

Alexandre Bois

Computer Vision and Pattern Recognition

A generic method for detecting anomalies on manufactured parts

This thesis investigates learning-based methods for industrial optical quality control. It provides autonomous solutions that do not require a computer vision expert to adapt to a new object. In the first part, we add an image normalization first…

Aitor Artola

Analytical chemistry

An Electrolyte-Gated Transistor combined with CRISPR/Cas13a for RNA detection

Ribonucleic acids (RNA) detection is a crucial step in the identification of viral or bacterial infections in humans and animals. To date, reverse transcriptase-polymerase chain reaction (RT-PCR) remains the gold standard, but it is based on an…

Benoît Piro, Pierre Guermonprez, Louis Renaud, Pierre Nioche, Eric Krejci, Aurelie Alleaume-Butaux, Nicolas Battaglini, Sébastien Sanaur

MADICA 2024 - 2024

Mathematics

Leveraging domain adaptation methods for federated learning applied to 2D mammography image classification

Breast cancer is the most commonly diagnosed cancer among women, with early detection significantly improving survival rates. X-ray imaging techniques are crucial in detecting, diagnosing, and monitoring breast cancer treatment. Deep Learning-based…

Gonzalo Iñaki Quintana

Analysis of PDEs

Development and assessment of physically informed learning methods : towards to multi-physical simulation in industrial contexts

Physics-informed deep learning approaches have gained huge attention in various engineering fields for the last few years thanks to the capability of incorporating physical knowledge, which is often represented using Partial Differential Equations …

Thi-Nguyen-Khoa Nguyen